MicrostockGroup
Agency Based Discussion => Adobe Stock => Topic started by: Artist on June 13, 2023, 00:26
-
People are flooding with AI content. No wonder if you are seeing downfall in your sales.
https://stock.adobe.com/in/search?filters%5Bcontent_type%3Aphoto%5D=1&filters%5Bcontent_type%3Aillustration%5D=1&filters%5Bcontent_type%3Azip_vector%5D=1&filters%5Bcontent_type%3Avideo%5D=1&filters%5Bcontent_type%3Atemplate%5D=1&filters%5Bcontent_type%3A3d%5D=1&filters%5Bcontent_type%3Aaudio%5D=0&filters%5Binclude_stock_enterprise%5D=0&filters%5Bis_editorial%5D=0&filters%5Bfree_collection%5D=0&filters%5Bcontent_type%3Aimage%5D=1&filters%5Bgentech%5D=only&order=creation&safe_search=1&limit=100&search_type=filter-select&search_page=1&get_facets=0
-
Topic: 9 Million+ AI generated photos
10 Millions+ now.
This is the end
Hold your breath and count to ten
Feel the earth move and then
Hear my heart burst again
For this is the end
I've drowned and dreamed this moment
So overdue, I owe them
Swept away, I'm stolen
Let the sky fall, when it crumbles...
-
A lot of it is very good, but a lot of it also rubbish. The fact that AS has been accepting AI content with all kinds of flaws since day one really makes me think it is a way of training their AI on the same data as the other apps without the legal implications. They just want as big a collection as quickly as possible. Contributors indemnify them by uploading and signing off that they have rights over the images. Adobestock then gets to claim they are ethical because they only use "their own" collection.
EDIT to add: I mean what lol:
https://stock.adobe.com/in/images/health-and-healthy-lifestyle-concept-man-riding-a-bicycle-with-a-bunch-of-fruit-generative-ai/599130317?
-
I don’t understand Adobe stock strategy… Real human artists goodbye or welcome?
-
I don’t understand Adobe stock strategy… Real human artists goodbye or welcome?
They look away from anything that doesn't concern their financial interest, universal business rule.
-
I don’t understand Adobe stock strategy… Real human artists goodbye or welcome?
They want to do away with having to pay artists/ photographers.
People will just pay to use the their AI, either online or integrated with their software via subscription. Their software will target the same market as Canva i.e. "designers" and "digital artists" as a distinct category will mostly disappear. Businesses/ marketing types/ advertisers/ entrepreneurs will be able to do the work via the apps directly with simple prompts and very basic skills. Much bigger market for them too.
-
I don’t understand Adobe stock strategy… Real human artists goodbye or welcome?
They want to do away with having to pay artists/ photographers.
People will just pay to use the their AI, either online or integrated with their software via subscription. Their software will target the same market as Canva i.e. "designers" and "digital artists" as a distinct category will mostly disappear. Businesses/ marketing types/ advertisers/ entrepreneurs will be able to do the work via the apps directly with simple prompts and very basic skills. Much bigger market for them too.
"They want to do away with having to pay artists/ photographers."
I don't think it's that simple. If they did that, who'd pay for the majority of their software that is their main bread and butter? The designers and photographers would be out of work so wouldn't need it.
There's a lot of AI imagery that is good, a lot more that is poor but one thing they all have in common (for me anyway), I can tell they're fake. There's just something false about a lot of them. Now, will clients want to use imagery that looks fake all the time. Everything has a certain style/look. I think the novelty factor will wear off at some point and the balance between real and AI imagery will settle down.
-
The only sensible thing to do is not waste any more time adding metadata to "real" images and submitting them to stock sites (especially adobe) and complaining about how they aren't getting reviewed or rejected. The big players are telling you in no uncertain terms that they don't need your images and it's foolhardy to expect to make any money off photography in the future.
-
The only sensible thing to do is not waste any more time adding metadata to "real" images and submitting them to stock sites (especially adobe) and complaining about how they aren't getting reviewed or rejected. The big players are telling you in no uncertain terms that they don't need your images and it's foolhardy to expect to make any money off photography in the future.
Really? Ok
-
I don't think it's that simple. If they did that, who'd pay for the majority of their software that is their main bread and butter? The designers and photographers would be out of work so wouldn't need it.
Adobe's entire AI pivot is to attract big money from enterprise subs so those corporates wouldn't have to hire many pesky little designers and artists and just get a manager to make those images easily with text prompts. Even now, they make most of their money from corporate subs anyway and losing individual designers/photographers would just be a tiny little bump on the road.
-
I don't think it's that simple. If they did that, who'd pay for the majority of their software that is their main bread and butter? The designers and photographers would be out of work so wouldn't need it.
That's just it. I think they want to pivot to making software that anyone can use, not just designers and photographers.
...one thing they all have in common (for me anyway), I can tell they're fake. There's just something false about a lot of them...
Have to disagree with this one too. The best of it is very high quality and I believe indistinguishable from other content. There is a lot of garbage out there too. Some of the apps themselves are not good, but the good ones very good, and improving incredibly fast.
-
I don't think it's that simple. If they did that, who'd pay for the majority of their software that is their main bread and butter? The designers and photographers would be out of work so wouldn't need it.
That's just it. I think they want to pivot to making software that anyone can use, not just designers and photographers.
...one thing they all have in common (for me anyway), I can tell they're fake. There's just something false about a lot of them...
Have to disagree with this one too. The best of it is very high quality and I believe indistinguishable from other content. There is a lot of garbage out there too. Some of the apps themselves are not good, but the good ones very good, and improving incredibly fast.
What if Adobe believes so strong in AI that they don't even care anymore about their core products?
-
They look away from anything that doesn't concern their financial interest, universal business rule.
[/quote]
Exactamundo!!!
-
We are at the begin of AI Age
-
A lot of it is very good, but a lot of it also rubbish. The fact that AS has been accepting AI content with all kinds of flaws since day one really makes me think it is a way of training their AI on the same data as the other apps without the legal implications. They just want as big a collection as quickly as possible. Contributors indemnify them by uploading and signing off that they have rights over the images. Adobestock then gets to claim they are ethical because they only use "their own" collection.
EDIT to add: I mean what lol:
https://stock.adobe.com/in/images/health-and-healthy-lifestyle-concept-man-riding-a-bicycle-with-a-bunch-of-fruit-generative-ai/599130317?
Wow that is really strange and wrong in so many ways. Should I ask, what's that white seahorse thing on the pumpkin (which isn't a fruit) neither are most of the other gourds and things, some which aren't healthy edible food. But ignoring nit picking, that's not a man! Well maybe. This is the open world of being any sex you want to be. But I mean by traditional standards, that's a female. Is that a bird on the white grapes. Do you steer a bicycle with a joystick in one hand. And one of my favorite errors in AI image, they cut her head off!
Nice find, that's really horrific. There's also a serious question about how she is "mounted" on that seat. ;D Ouch! I shouldn't ask about where the seat post is going?
(https://i.postimg.cc/d0ZKVbNQ/bicycle-butt.jpg)
-
A lot of it is very good, but a lot of it also rubbish. The fact that AS has been accepting AI content with all kinds of flaws since day one really makes me think it is a way of training their AI on the same data as the other apps without the legal implications. They just want as big a collection as quickly as possible. Contributors indemnify them by uploading and signing off that they have rights over the images. Adobestock then gets to claim they are ethical because they only use "their own" collection.
EDIT to add: I mean what lol:
https://stock.adobe.com/in/images/health-and-healthy-lifestyle-concept-man-riding-a-bicycle-with-a-bunch-of-fruit-generative-ai/599130317?
Wow that is really strange and wrong in so many ways. Should I ask, what's that white seahorse thing on the pumpkin (which isn't a fruit) neither are most of the other gourds and things, some which aren't healthy edible food. But ignoring nit picking, that's not a man! Well maybe. This is the open world of being any sex you want to be. But I mean by traditional standards, that's a female. Is that a bird on the white grapes. Do you steer a bicycle with a joystick in one hand. And one of my favorite errors in AI image, they cut her head off!
Nice find, that's really horrific. There's also a serious question about how she is "mounted" on that seat. ;D Ouch! I shouldn't ask about where the seat post is going?
(https://i.postimg.cc/d0ZKVbNQ/bicycle-butt.jpg)
That is what people like these days. Confusing and contradictive concepts. I think it totally fits in today's nihilism.
-
EDIT to add: I mean what lol:
https://stock.adobe.com/in/images/health-and-healthy-lifestyle-concept-man-riding-a-bicycle-with-a-bunch-of-fruit-generative-ai/599130317?
That now generates a 404 error - perhaps the crack inspection team was embarrassed at what they had allowed into the collection now someone pointed it out?
-
was searching for some of my ai on AS --- for ai generated with keywords "venice gondola" not only did mine not show up in first few pages (not a big concern here) but the first page & a half were filled with similars most of which had at best only tiny gondolas and none showed venice (more likely amalfi coast)
heres's the search https://tinyurl.com/3z35792b (https://tinyurl.com/3z35792b)
-
It's now 10+ million. I spent a few minutes looking though Gen AI work sorted by most recent. There are a few interesting and potentially useful items. There's a fair bit that's pretty to look at but I have no clue what its use case is as a stock image. The bulk of it looked repetitive, derivative and substandard to me. Some had so many similars I can't see why they were all approved even if one or two would make sense.
We need Gen AI water splashes, or hamburgers with fries, isolated strawberries or spoons of white sugar?? Tons of plastic looking women, robot/borg heads and roses on a wood background. Thousands of fantasy landscapes marked as "game backgrounds". Kitchen sinks full to the brim with peppers and vegetables, glass of seltzer, etc. etc.
If this is "quality", it's using a very odd measuring device
And AI is as hopeless at insect limbs as it is with human ones :)
-
was searching for some of my ai on AS --- for ai generated with keywords "venice gondola" not only did mine not show up in first few pages (not a big concern here) but the first page & a half were filled with similars most of which had at best only tiny gondolas and none showed venice (more likely amalfi coast)
heres's the search https://tinyurl.com/3z35792b (https://tinyurl.com/3z35792b)
The results get a lot better when you switch from "Most Recent" to "Relevance" and go to the first page. The reality still beats AI by a fair margin, though.
-
EDIT to add: I mean what lol:
https://stock.adobe.com/in/images/health-and-healthy-lifestyle-concept-man-riding-a-bicycle-with-a-bunch-of-fruit-generative-ai/599130317?
That now generates a 404 error - perhaps the crack inspection team was embarrassed at what they had allowed into the collection now someone pointed it out?
Funny. That was just a random pick I spotted right away. I could have picked from hundreds of examples!
-
Stock photography is coming to an end has been the most consistent topic on this forum for the last 15 years. And yet, here we are.
-
Stock photography is coming to an end has been the most consistent topic on this forum for the last 15 years. And yet, here we are.
You should see facts. Such technology was never build and this AI is something which can create 100s or professional works within minutes.
And it is self improving and learning program so it will become more accurate with time.
-
If you look around the unofficial FB groups they're full of posters asking "How many AIs can i submit at once without account getting blocked" and "Will i get caught if i dont flag as AI".
Huge numbers from India/Pakistan but not exclusively who may or may not be coordinated but certainly their strategy is literally create as much AI as possible and just shovel it in.
Anecdotally, a LOT is being submitted as normal RF and not being tagged as such, deliberately.
-
I think the quality that is coming in, is perfectly fine. I am also not worried about the similars and the duplicates, because all agencies have algorithms that deal with that. Only a fraction of the series is shown in the relevance search.
What worries me is the slowdown in inspections, which will get a lot worse when firefly goes live. It will be the first really easy to use app coming from a large company and stock agency. With an ethical training system. A lot of people who have not used ai before, will then start doing it.
The flood will increase 1000% or even more witzh all the youtubers screaming how easy it is to make money with ai on Adobe.
Which is why I hope they will inplement sensible upload limits, coupled to the sales success of a portfolio. Maybe even with a factor sales/portfolio size.
And look into their photo inspections, to fix what is broken.
But ai is here to stay, if Adobe has 100 million ai files, in addition to everything else, it is a competitive advantage because so far the other large agencies still don‘t have ai content.
And so far I am not seeing a slowdown in photo sales, but nothing that I photograph looks like what is being produced for ai.
eta: I am wondering if the ai collections looks like 90% repetitive midjourney not necessarily because this is what people upload, but because the style directions for the ai collection are heavily focussed on midjourney. Unlike photography or normal illustrations, maybe for gen ai they have "style guide".
Again, I am sure Adobe is doing what they do for maximal commercial gain. So if the midjourney look is what they or the customers prefer, then maybe we will have to adapt.
-
It takes time before new images emerge from the batch, are visible and therefore purchased, because of the algorithms. But once the inertia is overcome, the huge AI army will inevitably do huge damage to sales.
-
Almost everything looks very similar, like it's gone through the same "art" filter. I also see a lot weird shapes and artifacts that have no logic being there and that would normally disqualify the image but apparently the submission bar is set lower for AI. The whole thing is very discouraging. I could get on the bandwagon and have already tested a local SD iteration but every time I try to "create" something I feel like I'm cheating. Even when I get to something acceptable I just can't own the final result. I couldn't, even for a second, consider this thing mine. I can't imagine doing this for a living.
My take is this whole trend makes art accessible for the really bottom of the barrel, completely talentless and skill-less part of the crowd. Coincidentally these people are so happy to finally see something beautiful come as result of their efforts, they have no scrupules or second thoughts about owning it. As for the agencies, their moto has always been "Anything for a profit"
-
Personally I find the quirks and imperfections very interesting. It feels like the ai is striking back and making it their own as opposed to the prompters creation.
With a buyers hat on, I am very grateful that Adobe is offering a huge gen ai collection. I have asked many friends with businesses, they have all tried prompting and been more than disappointed with the results.
They have no time for that and will happily buy from an agency that offers a wide selection.
That includes people who have a good basis in photography or art, it simply takes much too long to learn the right prompts to get the result you want.
I also don‘t think that the „masses have no talent“. They aree simply not trained in art, like they might not be trained in music or metalwork or sky diving.
But everyone can learn to create, like everyone can learn to do photography.
Ai just speeds up the learning process, because you just have to focus on the end result and don‘t have to learn the individual technique like oil painting, watercolor, papercut or photography.
-
Almost everything looks very similar, like it's gone through the same "art" filter.
Either many people are purposefully going for the same effect, because it is popular, or they do not know that they can influence the look of an image by describing for example light conditions and you can achieve images that look "artsy" and overprocessed or images that look like more like candid photos, but I think there is some great diversity there, especially since not everyone is using the same AI generator.
-
Personally I find the quirks and imperfections very interesting. It feels like the ai is striking back and making it their own as opposed to the prompters creation.
With a buyers hat on, I am very grateful that Adobe is offering a huge gen ai collection. I have asked many friends with businesses, they have all tried prompting and been more than disappointed with the results.
They have no time for that and will happily buy from an agency that offers a wide selection.
That includes people who have a good basis in photography or art, it simply takes much too long to learn the right prompts to get the result you want.
I also don‘t think that the „masses have no talent“. They aree simply not trained in art, like they might not be trained in music or metalwork or sky diving.
But everyone can learn to create, like everyone can learn to do photography.
Ai just speeds up the learning process, because you just have to focus on the end result and don‘t have to learn the individual technique like oil painting, watercolor, papercut or photography.
I'm sure it takes certain amount of talent to find and push the "generate" button but to make the debate shorter just take a look of what these people did before their AI content. There's an abyss between before and after
Almost everything looks very similar, like it's gone through the same "art" filter.
Either many people are purposefully going for the same effect, because it is popular, or they do not know that they can influence the look of an image by describing for example light conditions and you can achieve images that look "artsy" and overprocessed or images that look like more like candid photos, but I think there is some great diversity there, especially since not everyone is using the same AI generator.
These images share at least half of their prompts. UHD, 8K, HDR, Trending on Artstation, High definition, Ultra detailed etc...
-
I'm sure it takes certain amount of talent to find and push the "generate" button but to make the debate shorter just take a look of what these people did before their AI content. There's an abyss between before and after
It is funny watching self professed “prompt gurus” write essays on the Midjourney Discord with incredibly flowery prompts and get results indistinguishable from the next person who uses a couple of words.
As long as you can write a sentence, take a few minutes to look at what other people have used and spend a bit of time learning the basic commands you can tailor results as much as anyone. You can produce results as good as anyone else right away, and tailor the results to whatever style want after an afternoon's practice.
If this wasn’t the case it would be a huge failure of Midjourney. The whole point of it is to allow people get results with basic prompts. And lets not forget you can now feed it images and get a descriptive prompt, reverse engineering a prompt to get similar results.
-
I'm sure it takes certain amount of talent to find and push the "generate" button but to make the debate shorter just take a look of what these people did before their AI content. There's an abyss between before and after
It is funny watching self professed “prompt gurus” write essays on the Midjourney Discord with incredibly flowery prompts and get results indistinguishable from the next person who uses a couple of words.
As long as you can write a sentence, take a few minutes to look at what other people have used and spend a bit of time learning the basic commands you can tailor results as much as anyone. You can produce results as good as anyone else right away, and tailor the results to whatever style want after an afternoon's practice.
If this wasn’t the case it would be a huge failure of Midjourney. The whole point of it is to allow people get results with basic prompts. And lets not forget you can now feed it images and get a descriptive prompt, reverse engineering a prompt to get similar results.
Yes, it's like eating vomit, then vomiting again. Reverse vomiting produces a new vomit ;D oh sorry for my particular humour which may offend these self-proclaimed artist teleprompters, which before the use of AI only produced visual crap. They now think they are Picasso. It also reminds me of modern music, where the sound produced by the remixed singer voice evokes in me the idea of digital vomit.
I think this metaphor is the correct one. It can occur following a massive ingestion, and it arises expeditiously. It is not very appetizing and is particularly repellent. But when the sheeps have nothing but vomit to eat, they won't find it too bad.
Imagine the dunce of the class who discovers chatGPT, uses it, proudly declares: "I wrote it myself", and who ends up being convinced of it himself.
It's very hard to get rid of vomAIt...
-
I completely get that "Quality" is a very subjective measure, and that all the agencies have the right to set their own standards for what they will and won't accept, but as I look through the 10+ million (acknowledged) genAI images in the Adobe Stock collection, I think there's work that should be rejected because it's faulty reality - not artistic point of view, not fantasy, not creative expression, but just a mistake.
I've been collecting examples - roast turkeys with four drumsticks and two wings; a lobster with three claws (the one in the front sort of circular); an office chair with three arms, one in the middle so you can't sit; a cow with a single bump-like udder in the middle of its belly and another with several balloon-like udders in the right (ish) place; an office chair with two legs and one arm that couldn't even stand up; a refrigerator with open doors too wide to close; a photo of London in the 1940s which not only isn't but is mismatched building bits, US flag with too many/few stars, stripes, stars that are squares or triangles, spiders with 10, 12, 14 legs and cockroaches with 8 . . .
Two issues with this type of thing from my perspective. One is that this type of content is defective - noise free, in focus, good color, but just a mistake.
The other is that some buyers may not realize the content is erroneous and just assume the agency has their back and it's good to use. An example: a buyer has not been lucky enough to eat lobster or even know what it looks like, but has been told to download a picture of a lobster meal. GenAI content is included in search results by default - you have to turn it off if you don't want it, so the unlucky buyer downloads a mutant genAI image with extra claws without realizing it's pretty but wrong. Doesn't an agency have some responsibility to weed out this type of "mistaken" content?
-
I completely get that "Quality" is a very subjective measure, and that all the agencies have the right to set their own standards for what they will and won't accept, but as I look through the 10+ million (acknowledged) genAI images in the Adobe Stock collection, I think there's work that should be rejected because it's faulty reality...
Exactly what I was talking about. There's loads of stuff that they would never even think about accepting if it was the result of dodgy photoshop work, but they've been letting that garbage in from day one if it is AI generated. Very odd policy.
-
Like I said, the 8 fingers, 3 claws etc…IMO are „typical“ of ai. Which why for me it makes perfect sense to take it.
Otherwise it could just be a normal illustration or photoshop work.
-
AI images all look great at a small thumbnail size BUT 90% unacceptable even at 2000px.
Blurred, out of focus, strange image details..
And most of the ones I've seen so far have been upscaled to 5000+ pixels.
Customer refunds foreseeable?
Also found the information about this during the upload if you mark your images as AI, "Most common rejection reason for AI-generated images ...an image that is 1000 pixels long should not be scaled up to 8000 pixels..." (Indivstock)
-
Doesn't an agency have some responsibility to weed out this type of "mistaken" content?
In any case, the agency makes no effort to do so. Probably also related to the automation of validation process.
As I already said in antother thread, I think it's our relationship to real, natural, logical, truth, beauty, memory, history that is challenged.
-
The other is that some buyers may not realize the content is erroneous and just assume the agency has their back and it's good to use. An example: a buyer has not been lucky enough to eat lobster or even know what it looks like, but has been told to download a picture of a lobster meal. GenAI content is included in search results by default - you have to turn it off if you don't want it, so the unlucky buyer downloads a mutant genAI image with extra claws without realizing it's pretty but wrong. Doesn't an agency have some responsibility to weed out this type of "mistaken" content?
Maybe no more than they have the responsibility to make sure captions and keywords are correct, for a similar reason (e.g. the junior has been tasked with finding images, and doesn't know any better)
-
Like I said, the 8 fingers, 3 claws etc…IMO are „typical“ of ai. Which why for me it makes perfect sense to take it.
How, in any universe, is this a useful image of a roast turkey?
(https://as2.ftcdn.net/v2/jpg/05/43/05/23/1000_F_543052358_bq0lwpGBJlP1G82GXjYtclUe4TfSIjjf.jpg) (https://stock.adobe.com/images/christmas-turkey/543052358)
Or this of a boiled lobster?
(https://as2.ftcdn.net/v2/jpg/05/84/30/09/1000_F_584300917_fFHTPh5WC3htFy6CDNFV9LGVqd2jRjbG.jpg) (https://stock.adobe.com/images/lobster-with-lemon/584300917)
There are lots and lots of this type of "mistakes of AI" - I'm not just cherry picking a few with flaws.
Or, if this type of could-easily-be-mistaken-for-real content is really important, then slap an overlay on them, like "EDITORIAL ONLY" or "PREMIUM" have. Otherwise it's a fast ticket to getting fired from your job for being mocked on social media instead of promoting whatever it was you were supposed to be pitching.
-
If this is "quality", it's using a very odd measuring device
And AI is as hopeless at insect limbs as it is with human ones :)
perhaps a spider - cockroach hybrid?
-
If this is "quality", it's using a very odd measuring device
And AI is as hopeless at insect limbs as it is with human ones :)
perhaps a spider - cockroach hybrid?
actually the turkey is a creation of John Madden for holiday football games
and i'd love to find a lobster w claws that size!
-
Stock photography is coming to an end has been the most consistent topic on this forum for the last 15 years. And yet, here we are.
You should see facts. Such technology was never build and this AI is something which can create 100s or professional works within minutes.
And it is self improving and learning program so it will become more accurate with time.
which is it? AI creations are terrible or AI images are 'professional'
-
I completely get that "Quality" is a very subjective measure, and that all the agencies have the right to set their own standards for what they will and won't accept, but as I look through the 10+ million (acknowledged) genAI images in the Adobe Stock collection, I think there's work that should be rejected because it's faulty reality - not artistic point of view, not fantasy, not creative expression, but just a mistake.
I've been collecting examples - roast turkeys with four drumsticks and two wings; a lobster with three claws (the one in the front sort of circular); an office chair with three arms, one in the middle so you can't sit; a cow with a single bump-like udder in the middle of its belly and another with several balloon-like udders in the right (ish) place; an office chair with two legs and one arm that couldn't even stand up; a refrigerator with open doors too wide to close; a photo of London in the 1940s which not only isn't but is mismatched building bits...
Two issues with this type of thing from my perspective. One is that this type of content is defective - noise free, in focus, good color, but just a mistake.
The other is that some buyers may not realize the content is erroneous and just assume the agency has their back and it's good to use. An example: a buyer has not been lucky enough to eat lobster or even know what it looks like, but has been told to download a picture of a lobster meal. GenAI content is included in search results by default - you have to turn it off if you don't want it, so the unlucky buyer downloads a mutant genAI image with extra claws without realizing it's pretty but wrong. Doesn't an agency have some responsibility to weed out this type of "mistaken" content?
the really sad thing is not that AI produces these freaks, but (as others have noted) that people upload them & they're accepted. even my steampunk dinosaurs have the proper number of limbs!
-
@Jo Ann
The Lobster ist fantastic! I would use this and enjoy all the comments and interactions you can get from using that.
The turkey is less attractive for me but I am sure somebody will buy it because it was visibly done with ai.
I think Adobe should put together an edited ai collection to promote and pictures like the lobster should be the top promoted images.
-
Stock photography is coming to an end has been the most consistent topic on this forum for the last 15 years. And yet, here we are.
T
Stock photography is coming to an end has been the most consistent topic on this forum for the last 15 years. And yet, here we are.
You should see facts. Such technology was never build and this AI is something which can create 100s or professional works within minutes.
And it is self improving and learning program so it will become more accurate with time.
which is it? AI creations are terrible or AI images are 'professional'
the question is (now we are at the beginning of AI) if actual photos, vectors and/or videos will be replacedechnology was never build and this AI is something which can create 100s or professional works within minutes.
And it is self improving and learning program so it will become more accurate with time.
which is it? AI creations are terrible or AI images are 'professional'
The question is (and now we are just at the beginning with AI) if photos, vectors and videos will be replaced by AI or not.
The next question is, if not, are your photos, vectors or videos findable between this garbage?
-
The question is (and now we are just at the beginning with AI) if photos, vectors and videos will be replaced by AI or not.
The next question is, if not, are your photos, vectors or videos findable between this garbage?
SVH, everytime you post, I add +1 ;)
-
The Lobster ist fantastic! I would use this and enjoy all the comments and interactions you can get from using that.
The Lobster is fantasplastic!
There are lots and lots of this type of "mistakes of AI"...
AI IS mistake
People who take photographs never count the number of fingers, one would have to wonder why this should happen now.
getting fired from your job for being mocked on social media
Fantastic then! the opportunity for that person to go back to real organic and true life! ;)
-
The question is (and now we are just at the beginning with AI) if photos, vectors and videos will be replaced by AI or not.
The next question is, if not, are your photos, vectors or videos findable between this garbage?
SVH, everytime you post, I add +1 ;)
Renegades of the sheep community :)
-
was searching for some of my ai on AS --- for ai generated with keywords "venice gondola" not only did mine not show up in first few pages (not a big concern here) but the first page & a half were filled with similars most of which had at best only tiny gondolas and none showed venice (more likely amalfi coast)
heres's the search https://tinyurl.com/3z35792b (https://tinyurl.com/3z35792b)
These are not gondolas but a different type of ligurian boat for fishing. And the location is more similar to ligurian coast (5 Terre). But the problem is that there aren't valid verification of non AI images, if you switch with the same search in "non AI" you'll see images that probably are AI images, like the venice paintings or similar.
-
Actually I'm wondering, is all of this stuff Midjourney? The generated resolution is not very high so is it all upscaled to a higher resolution? Or are there certain paid plans or AI generators that produce higher resolutions, or higher quality results?
Not that I intend to use AI myself; it's an uphill battle, and it feels like cheating... Anyone who knows how to write prompts can do it, whereas not everybody can shoot high quality photos or videos or draw beautiful vector art.
About agencies needing to take responsibility to check these images for accuracy: that will probably never happen, because reviewers are trained to determine photo image quality (like noise, out of focus, etc), they're not biologists, historians or food experts. They won't check Wikipedia for species of insects to correctly determine the shape and size of legs, or the correct size or color of a piece of fruit.
-
Whenever I have used AI software half the results are full of artifacts and would never be accepted.
The next stage may wipe out stock agencies as customers will just use the simple software themselves to create images.
-
If you do a search on Adobe Stock for "generative" and select the filter to exclude Generative AI, you have 644,228 results. I spot checked a number of them and they (a) look like AI and (b) the title quite clearly says they were generative AI (obviously there will be some that have that search term for another reason, but it's still a large number of improperly identified images).
So these predated the filter, or someone forgot to check the box or something - and reviewers didn't notice or don't have a mandate to get that sort of thing fixed?
This isn't just stuff from a long time ago. The newest image in the collection at time of writing was 611697106 and the newest AI image that didn't have the GenAI tag set was 610830747
This must be corrected. Even Adobe's corporate spin at various public events has included the importance of being transparent - tagging AI generated images as such - and surely that has to include the Adobe Stock collection.
"As a trusted partner to individuals and businesses of all sizes, Adobe develops and deploys all AI capabilities with a customer-centric approach and according to its AI Ethics principles to ensure content and data transparency. Content Credentials provide “nutrition labels” for digital content and are a key pillar of Adobe’s AI principles."
https://news.adobe.com/news/news-details/2023/Adobe-Unveils-Future-of-Creative-Cloud-with-Generative-AI-as-a-Creative-Co-Pilot-in-Photoshop-default.aspx/default.aspx
https://contentauthenticity.org/
https://venturebeat.com/ai/adobe-commits-to-transparency-in-use-of-generative-ai/
"And how will we know whether something we see was created by a human or a computer?"
On Adobe Stock, one would hope you'd know because content was properly handled by the upload and inspection process, and ideally, thumbnails would be overlaid by a "Gen AI" box (like editorial and premium images have)
-
ai images are all upsized, usually with Topaz software. They are also often heavily edited in photoshop, because they don‘t come out perfect.
-
There are multiple thousands of seriously wrong flags in the GenAI collection, particularly US flags, and with Independence Day near, wouldn't it be great if the bogus flags could be disappeared? There are over half a million human-created ones that are fine to use.
And perhaps the "quality" reviewers could be given the rules for flags so they don't let more of this stuff in. What possible advantage is there for Adobe Stock to have tens of thousands of mangled flags in the collection?
There are many examples, but here's just a handful:
(https://as2.ftcdn.net/v2/jpg/06/04/43/07/1000_F_604430750_ubhvZ241k7heFleouGsCUU9ZSY2UIL12.jpg) (https://stock.adobe.com/images/abstract-american-flag-4th-july-for-wallpaper-design-abstract-firework-festive-background-template-banner/604430750)
(https://as1.ftcdn.net/v2/jpg/06/03/96/74/1000_F_603967437_HT2s2fyxArmbp9AOm9qxXV6VlzguAaci.jpg) (https://stock.adobe.com/images/illustration-for-independence-day-of-the-united-states-july-4th-soldier-with-american-flag-under-a-beautiful-sun-generative-ai/603967437) This has been removed
(https://as1.ftcdn.net/v2/jpg/06/11/91/74/1000_F_611917435_nB9ele4KYstcuWrRzTm5oIRIHvngT8wr.jpg) (https://stock.adobe.com/images/soldiers-in-front-of-usa-flag-with-dark-background-generative-ai/611917435) This has been removed
(https://as1.ftcdn.net/v2/jpg/06/03/65/46/1000_F_603654658_mp3b12yYMfJjeACL1noFRwZB13NRsK8I.jpg) (https://stock.adobe.com/images/free-photo-usa-america-national-flag-isolated-3d-white-background-made-with-generative-ai/603654658)
-
Like I said, the 8 fingers, 3 claws etc…IMO are „typical“ of ai. Which why for me it makes perfect sense to take it.
How, in any universe, is this a useful image of a roast turkey?
(https://as2.ftcdn.net/v2/jpg/05/43/05/23/1000_F_543052358_bq0lwpGBJlP1G82GXjYtclUe4TfSIjjf.jpg) (https://stock.adobe.com/images/christmas-turkey/543052358)
Or this of a boiled lobster?
(https://as2.ftcdn.net/v2/jpg/05/84/30/09/1000_F_584300917_fFHTPh5WC3htFy6CDNFV9LGVqd2jRjbG.jpg) (https://stock.adobe.com/images/lobster-with-lemon/584300917)
There are lots and lots of this type of "mistakes of AI" - I'm not just cherry picking a few with flaws.
Or, if this type of could-easily-be-mistaken-for-real content is really important, then slap an overlay on them, like "EDITORIAL ONLY" or "PREMIUM" have. Otherwise it's a fast ticket to getting fired from your job for being mocked on social media instead of promoting whatever it was you were supposed to be pitching.
Good points. I believe AI will "get there" not making "mistakes", but it may take long enough to give AI images a really bad name among commercial users...enough so that they just avoid it and defer to using actual photographic imagery for quite some time.
-
I think the flags are ok, because the customer can immediately see what the problem is. If they don't like it, they won't download.
My biggest gripe are with the isolations and everything that is difficult to see and that you will only notice after downloading.
Ai content is clearly labeled, so the customers know that there might be some problems.
But most important: I am comfortable if they take millions of crappy files as long as they don't decline my beautiful perfect ai creations!!
:)
-
I think the flags are ok, because the customer can immediately see what the problem is. If they don't like it, they won't download.
Maybe... if buyer is buying some other country flag, not his own, he won't notice immediately something is wrong with it.
-
True, but basically the ai label is also a warning sign. Our customers are professionals.
But perhaps they could make the ai label larger or more visible and add a few more articles or blog posts how ai images are a computer being creative and not documentary photography. Similar to how a chat gpt text is not a scientific paper but has to be fact checked.
They can write the articles with a positive vibe, personally I would put the ai mistakes like the lobster front and center. These images will lead to endless engagements on social media.
-
Even Adobe's corporate spin at various public events has included the importance of being transparent - tagging AI generated images as such - and surely that has to include the Adobe Stock collection.
"As a trusted partner to individuals and businesses of all sizes, Adobe develops and deploys all AI capabilities with a customer-centric approach and according to its AI Ethics principles to ensure content and data transparency. Content Credentials provide “nutrition labels” for digital content and are a key pillar of Adobe’s AI principles."
Adobe also tells its customers they can "trust" Adobe because they compensated their artists for the use of the database ...
Even if they "plan" to do this in the near/far future, it is/was not true when they launched Firefly and started accepting AI images as stock.
-
Whether they like it or not, they've already got a *lot* of AI images sold as stock and unlabelled.
-
Whether they like it or not, they've already got a *lot* of AI images sold as stock and unlabelled.
I've seen a lot like that too. And from the Adobe Forums, someone pointed out this portfolio which appears to be AI created but isn't labeled in the title or with the Generative AI tag...
https://stock.adobe.com/contributor/211097879/hera-kim
The post was made at the end of April and the portfolio is still there, still not labeled as AI (and no response to the post either).
In the same thread, Mat Hayward had posted 2 days earlier "We have done a deep analysis on photos and illustrations not tagged by contributors as generative AI and have included those detected as AI in the filter. It should be much better for those wishing to eliminated generative AI content from their search results. We are constantly working to improve the Adobe Stock experience for everyone!"
I did a check on one of the images from the hera-kim port (the crazy ornate wedding venues) and it does not show up with "Generative AI Only" and does with "Exclude Generative AI". I obviously don't know if the person who posted this knows how the images were created, but some of the "mistakes" look very much like AI generated mistakes to me
-
That looks like a midjourney port.
-
If you trawl the FB groups for example there's one recent guy who (on SS admittedly) submitted some woeful mobile snapshots of his garden to get "approved".
Badly composed, awfully exposed garden growth. 7 of them.
Next day post is "Please visit and buy my portfolio" and the link now contains some 30+ images of clearly AI generated cups of coffee in addition to those weeds.
And yep, he's listed as AS contributor too and far from alone.
-
Whether they like it or not, they've already got a *lot* of AI images sold as stock and unlabelled.
I've seen a lot like that too. And from the Adobe Forums, someone pointed out this portfolio which appears to be AI created but isn't labeled in the title or with the Generative AI tag...
https://stock.adobe.com/contributor/211097879/hera-kim
The post was made at the end of April and the portfolio is still there, still not labeled as AI (and no response to the post either).
The portfolio is gone.
-
I think nothing will truly replace the spirit and the soul of a human without seeming like a fraud somehow eventually!
-
...The portfolio is gone.
These portfolios don't need to go away, just get tagged as AI - perhaps it'll come back after tagging??
Perhaps the same process/magic can be worked on this portfolio which appears (to me; based on style and all the mistakes) to be 100% Gen AI but isn't tagged as such
https://stock.adobe.com/search?creator_id=211433618
It has levitating ice cream cones, hot dog soup, gibberish text, mangled cutlery, strange laptop keyboards - all the hits :)
-
...The portfolio is gone.
These portfolios don't need to go away, just get tagged as AI - perhaps it'll come back after tagging??
Perhaps the same process/magic can be worked on this portfolio which appears (to me; based on style and all the mistakes) to be 100% Gen AI but isn't tagged as such
https://stock.adobe.com/search?creator_id=211433618
It has levitating ice cream cones, hot dog soup, gibberish text, mangled cutlery, strange laptop keyboards - all the hits :)
AI doesn't get bicycle spokes or machinery right either.
-
Stock photography is coming to an end has been the most consistent topic on this forum for the last 15 years. And yet, here we are.
That and how you can't make money in microstock any more. My stock earnings since 2006:
-
Stock photography is coming to an end has been the most consistent topic on this forum for the last 15 years. And yet, here we are.
That and how you can't make money in microstock any more. My stock earnings since 2006:
looks nice if.. , what grid numbers are here missing? in 10 or 100 USD steps?
-
AI doesn't get bicycle spokes or machinery right either.
I tried it on marine life and underwater (because thats mostly what i do myself).
Even entering pretty specific fish and species names most of the outputs are wrong to the extent you either think "what is that" or have invented new species previously unknown to science.
I did get a 10 arm octopus spat out once. And one with a claw (only 1 though).
-
Stock photography is coming to an end has been the most consistent topic on this forum for the last 15 years. And yet, here we are.
That and how you can't make money in microstock any more. My stock earnings since 2006:
For everyone complaing that stock is dead, there is someone else who just uploads and keeps making more money.
There are so many gaps in all the agency collections, endless possibilties.
And so many agencies to work with. Personally exploring the smaller players and uploading older content. It makes money. Even 18 year old files can bring income in the right new environment.
But at least for me, Adobe is currently the most interesting.
-
The genAI collection at Adobe Stock continues to grow - it's now just over 10.75 million.
I've continued to look at what's being added; the total fantasy stuff is visually rich, although the market for it may not be all that large (all the video game backgrounds look pretty, but will video game companies actually be able to use them instead of making their own?).
The images that attempt to portray the real world (including specific places, something the written rules explicitly forbid) are largely terrible. Here are just two examples (click to see detail page):
(https://t3.ftcdn.net/jpg/06/06/51/74/240_F_606517433_pv0p7PCBK1ZlWNasBN4RaE2Vg0ylX1O6.jpg) (https://stock.adobe.com/images/group-of-young-people-having-fun-sitting-around-the-campfire-generative-ai/606517433)
(https://t3.ftcdn.net/jpg/06/02/33/80/240_F_602338002_6BP7Qt3C5ebm0Kb4FPnZ2fHhh6BwFwCX.jpg) (https://stock.adobe.com/images/female-caucasian-50s-cooking-a-meal-in-kitchen-generative-ai-aig23/602338002)
And here's something where the terrible quality of the output should have resulted in a rejection (and there are others equally awful in the same portfolio)
(https://t3.ftcdn.net/jpg/06/14/43/14/240_F_614431429_688epFrdWqMMshgjp77izZO0zeZvoDMj.jpg) (https://stock.adobe.com/images/old-town-of-brugge-in-belgium/614431429)
Large numbers of similars (from a single contributor) keep appearing - 36 brick wall backgrounds, over 100 beach sunset with palm trees, 50 hamburgers with exploding food bits, over 1,000 abstract wave line backgrounds, 58 whole peach backgrounds . . .
It certainly appears the goal is to bulk up this part of the collection as quickly as possible and whatever criteria are in place are not at all the same as for the "regular" collection.
-
The collection is now over 11 million gen AI images (there are more but they're not marked. I can't know which the 11 millionth is, but how about this one of a waterslide no one should ever go on:
(https://t3.ftcdn.net/jpg/06/07/78/26/240_F_607782683_D2swCUe8BluvzjreN4s29qOAmqVjVZ83.jpg) (https://stock.adobe.com/images/water-slide-with-cascading-waterfall-at-water-park-created-with-generative-ai/607782683)
-
..no one should ever go on:
(https://t3.ftcdn.net/jpg/06/07/78/26/240_F_607782683_D2swCUe8BluvzjreN4s29qOAmqVjVZ83.jpg) (https://stock.adobe.com/images/water-slide-with-cascading-waterfall-at-water-park-created-with-generative-ai/607782683)
lol i love it, i saw 2 deadly endings already, how many can u find?
-
Won't put the in this thread as it'll detail it but yesterday i decided to try with generative fill on my own images.
Summary:- we stayed up for 5 hours and haven't laughed as much in years.
Literally not a single suggestion worked for inserting specific animals, created some horrific species previously unknown to science.
-
Adobe Stock has a Discord server for contributors and it now includes a "Quality-Control" channel where contributors can post links to images that should not have been accepted in the first place, and these will be passed on to moderation for re-consideration.
-
That's a nice idea, and I have thousands of examples I could offer, but I'm not allowed to post in that channel. I'm not level 10 (whatever level of activity that means) so I can't participate.
"The rules:
1) We post the URL to the image, not to a portfolio. Links to a portfolio where we need to find the image will be removed, it's not helping us 🙂
2) We explain the issue we see (auto traced vector; weird hands, etc.).
3) Not everyone can participate on this channel. You need to be an Active Member (having reached level 10) or a Creative Challenge Winner (perks of winning).
4) Slowmodeis active, so this won't get flooded. Be patient, or try to put several links on the same message.
5) We will remove messages about quality issues on other chat channels.
6) If this becomes a space for vendettas or personal attacks, people engaging on those behaviors will be on time out for 48 hours.
7) I reserve the right to change/adapt/modify or add new rules here, that won't be applied retroactively."
I get that they don't want spam or nuisance posts, but requiring you be active in their Discord channel versus licensing lots of stock images seems like the wrong way to go about it...
-
Is this work well paid by Adobe? ::) ;)
-
That's a nice idea, and I have thousands of examples I could offer, but I'm not allowed to post in that channel. I'm not level 10 (whatever level of activity that means) so I can't participate.
I posted the deadly waterslide one for you!
-
Adobe Stock has a Discord server for contributors and it now includes a "Quality-Control" channel where contributors can post links to images that should not have been accepted in the first place, and these will be passed on to moderation for re-consideration.
When will "Quality-Control" channel where contributors could inspect the bin where real photographies should not have been placed in the first place, and these would be passed on to moderation for re-consideration?
-
Facebook spammed but interesting.
Is that even allowed? GPT to Midjourney to upscale content and upload?
I cant find the guys portfolio but interested to see how much of his stuff is flagged as AI and how much is normal stock.
-
Facebook spammed but interesting.
Is that even allowed? GPT to Midjourney to upscale content and upload?
Yes, why should it not be allowed?
-
That depends if its flagged as AI content or not.
If not, its not allowed.
-
The collection is now over 11 million gen AI images (there are more but they're not marked. I can't know which the 11 millionth is, but how about this one of a waterslide no one should ever go on:
The death trap water slide has now been removed since I posted it on the Adobe Stock Discord :)
-
The collection is now over 11 million gen AI images (there are more but they're not marked. I can't know which the 11 millionth is, but how about this one of a waterslide no one should ever go on:
The death trap water slide has now been removed since I posted it on the Adobe Stock Discord :)
Thank you. I've noticed a few other images that have been "disappeared". I have watermarked previews or I'd think I had gone crazy :)
The problem continues though as recent approvals are still laden with "oops" images at a rapid rate. Adobe has to get a handle on this on the inspection side IMO.
-
And the collection (of acknowledged genAI content) is now over the 11.5million threshold - 11,507,885
July 12 it is 11,911,724
-
And the collection (of acknowledged genAI content) is now over the 11.5million threshold - 11,507,885
With the same pace, I predict AI content to cross 25-30 million mark by the end of this year.
-
When firefly opens for commercial stock use, it will accelerate drastically. Maybe even 100 million in 2 years.
I know a lot of producers holding back because of the legal issues but they will all jump into firefly once it is ready for commercial stock.
Other agencies will probably also offer properly licensed ai software. The Getty/Nvidia partnership could be interesting.
Wouldn't be surprised if Getty only allows content from their own ai.
-
Its what image farms are desperate for - can completely eliminate the slow process of actually obtaining images and now its just a question of computing power and staff to do it.
Literally goes into a data processing as opposed to image creation task.
-
Almost a month now since the initial message and the data is touching 12 million.
-
One month after the OP, the Adobe Stock genAI collection (tagged; there are more that are not) is over 12 million - 12,004,534
We can celebrate by making a yummy fruit smoothie - although I'm really afraid of this mixer...
(https://t4.ftcdn.net/jpg/06/00/43/27/240_F_600432797_iC6gVj94tCdTjaJ6rYnhxt8g4LO3GBrx.jpg) (https://stock.adobe.com/images/mixer-of-fruit-juice-on-the-wooden-table-ai-generative/600432797)
Whole fruits might be equally scary (same series from today's approvals)
(https://t3.ftcdn.net/jpg/06/00/43/28/240_F_600432800_kO1DhYnjL9aTo6diDjX9PQtZa2wGLMxP.jpg) (https://stock.adobe.com/images/mix-of-fruit-on-the-wooden-table-ai-generative/600432800)
I'll spare you the missing chair legs, stairs into the ceiling, chimps with extra arms, bizarre calculators/tools/planners etc.
After looking at recent acceptances I am in need of strong medicine - might this pill be big enough??
(https://t4.ftcdn.net/jpg/06/19/62/89/240_F_619628958_QyKuUJHBsZtYG9tsubKCqaCpYuLnmOAD.jpg) (https://stock.adobe.com/images/pills-and-receipts-on-the-table/619628958)
-
One month after the OP, the Adobe Stock genAI collection (tagged; there are more that are not) is over 12 million - 12,004,534
We can celebrate by making a yummy fruit smoothie - although I'm really afraid of this mixer...
(https://t4.ftcdn.net/jpg/06/00/43/27/240_F_600432797_iC6gVj94tCdTjaJ6rYnhxt8g4LO3GBrx.jpg) (https://stock.adobe.com/images/mixer-of-fruit-juice-on-the-wooden-table-ai-generative/600432797)
Whole fruits might be equally scary (same series from today's approvals)
(https://t3.ftcdn.net/jpg/06/00/43/28/240_F_600432800_kO1DhYnjL9aTo6diDjX9PQtZa2wGLMxP.jpg) (https://stock.adobe.com/images/mix-of-fruit-on-the-wooden-table-ai-generative/600432800)
I'll spare you the missing chair legs, stairs into the ceiling, chimps with extra arms, bizarre calculators/tools/planners etc.
After looking at recent acceptances I am in need of strong medicine - might this pill be big enough??
(https://t4.ftcdn.net/jpg/06/19/62/89/240_F_619628958_QyKuUJHBsZtYG9tsubKCqaCpYuLnmOAD.jpg) (https://stock.adobe.com/images/pills-and-receipts-on-the-table/619628958)
Oral medication has that prefix because it isnt the only route in. Some of those pills and many of the pills on the similar images offered have a distinctly rectal whiff about them.
-
Well this explains a lot. So I always look for niche areas I can exploit. This is one that I'll leave to others.
Testicular Cancer is on the rise and this is the offering for it with A.I. One image ...
Aside from the obvious complete lack of medical accuracy ... this is the description.
A.i. images are not reviewed. End of story and please don't abuse it ... 🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣 ... I'm sure it won't be abused.
-
The genAI collection is now 12,364,430 and growing. Looking at this morning's approvals, it continues to include images obviously sub-standard. Here are just a few to illustrate:
(https://t3.ftcdn.net/jpg/06/21/12/20/240_F_621122058_cluTmLl0U9eB4PYSZMP7JOlTrPwXillw.jpg) (https://stock.adobe.com/images/luxury-kitchen-decorating-ideas-ai/621122058)
(https://t4.ftcdn.net/jpg/06/20/77/41/240_F_620774114_IHVd0xLfocCBXpAFEd4oCPD7LtooGFl5.jpg) (https://stock.adobe.com/images/lake-and-mountains-generative-ai/620774114)
The title of this one is concept. The keywords give no useful clues:
cabinet office drawer file business computer furniture storage metal datum filing document file archive three-dimensional case box illustration folder technology icon document archive server equipment
(https://t3.ftcdn.net/jpg/05/97/81/88/240_F_597818811_Wgl1ky1q3FYNqgeeQhAZLrVYfuHbPtDV.jpg) (https://stock.adobe.com/images/concept/597818811)
(https://t3.ftcdn.net/jpg/06/19/26/30/240_F_619263000_5Mghk1KJT7kH9UmiJZMhfdi3lt5CQg5B.jpg) (https://stock.adobe.com/images/school-scenery-background-photo/619263000)
Wonky hands are still an issue
(https://t4.ftcdn.net/jpg/06/03/39/85/240_F_603398566_6Cif436YkSFtVjOTnEKGoVvvpLnEyaxY.jpg) (https://stock.adobe.com/images/person-preparing-salad-generative-ai/603398566)
(https://t3.ftcdn.net/jpg/06/13/83/44/240_F_613834455_LntJekOXcfES8ciC3P2i0BSIgYEZWCC6.jpg) (https://stock.adobe.com/images/a-group-of-energetic-young-athletes-playing-softball-in-a-sunny-stadium-kids-league-they-just-won-and-celebrating-it/613834455)
A favorite from yesterday: "beautiful nature wallpaper generated by AI tool". The keywords are so bad that anyone searching for hammock palm trees won't find it
(https://t3.ftcdn.net/jpg/06/21/47/84/240_F_621478408_UVFwECxCGQakg9kiMlghd29VTcNgejnG.jpg) (https://stock.adobe.com/images/beautiful-nature-wallpaper-generated-by-ai-tool/621478408)
One more from today - lobster is near and dear to my heart and this "lobster roll" is inedible! Those objects are not lobster claws or tails; even if they were, you need to take the shell off
(https://t4.ftcdn.net/jpg/06/20/75/11/240_F_620751146_6CrF9XxFU3ywfEZ6urFJLTMBYj8EVqRf.jpg) (https://stock.adobe.com/images/a-delectable-lobster-roll-featuring-chunks-of-succulent-lobster-meat-nestled-in-a-buttery-toasted-brioche-bun-served-with-a-side-of-crispy-fries/620751146)
Highroller in Portland does the best lobster rolls anywhere (just in case someone wants to compare the real thing with this imposter Adobe Stock has accepted :)
https://highrollerlobster.com/menu/
-
Highroller in Portland does the best lobster rolls anywhere (just in case someone wants to compare the real thing with this imposter Adobe Stock has accepted :)
them's fightin' words & you're cracking open a new debate...
My wife grew up on Mount Desert Island, Maine, & one our favorites, among several others, was
Thurston's Lobster Pound in Bernard
of course, we didnt go out too often as her brother was a lobsterman
-
Highroller is very recent (2015, I think). For boiled lobster I'd go for Chauncey Creek in Maine - given that I don't have any family members considerate enough to be in the lobster business.
And bear in mind I'm a first generation immigrant, so what do I know? :)
-
All agencies should clearly mention a warning just like shuttertsock do with their AI content.
Shutterstock does not review AI-generated content for compliance with Shutterstock’s content compliance standards. AI-generated content may not be used to infringe, misappropriate, or violate the intellectual property or other rights of any third party, to generate spam, false, misleading, deceptive, harmful, or violent imagery.
-
W-T-F Adobe? ? ? ? ?
two people shaking hands in front of a star filled sky. generative ai
(https://t4.ftcdn.net/jpg/06/03/38/99/240_F_603389941_S65PTJbl1YEkQ9mC13x3JxVXOW52Em8i.jpg) (https://stock.adobe.com/images/two-people-shaking-hands-in-front-of-a-star-filled-sky-generative-ai/603389941)
a couple of men shaking hands over a sunburst. generative ai
(https://t4.ftcdn.net/jpg/06/03/38/99/240_F_603389964_PwUf2hXtEW3EtvFdyYGhUgXvW5qQzTOI.jpg) (https://stock.adobe.com/images/a-couple-of-men-shaking-hands-over-a-sunburst-generative-ai/603389964)
two men shaking hands in front of a background with a pattern. generative ai
(https://t4.ftcdn.net/jpg/06/03/38/99/240_F_603389976_GxwQUr87krEx52xpCFm95rghPs9zXGpb.jpg) (https://stock.adobe.com/images/two-men-shaking-hands-in-front-of-a-background-with-a-pattern-generative-ai/603389976)
-
Now it's over the 12.5 million threshold - 12,512,192
-
edit: Nevermind ::)
-
Will this news affect those who make generative AI images for stock sites?
OpenAI, Google, others pledge to watermark AI content for safety -White House
https://www.reuters.com/technology/openai-google-others-pledge-watermark-ai-content-safety-white-house-2023-07-21/ (https://www.reuters.com/technology/openai-google-others-pledge-watermark-ai-content-safety-white-house-2023-07-21/)
-
This morning the collection has crossed the 13 million mark (13,006,207)...
-
One month after the OP, the Adobe Stock genAI collection (tagged; there are more that are not) is over 12 million - 12,004,534
We can celebrate by making a yummy fruit smoothie - although I'm really afraid of this mixer...
(https://t4.ftcdn.net/jpg/06/00/43/27/240_F_600432797_iC6gVj94tCdTjaJ6rYnhxt8g4LO3GBrx.jpg) (https://stock.adobe.com/images/mixer-of-fruit-juice-on-the-wooden-table-ai-generative/600432797)
The mixer is clearly Borg and in the process of assimilating the kitchen.
Resistance is futile!
-
One month after the OP, the Adobe Stock genAI collection (tagged; there are more that are not) is over 12 million - 12,004,534
We can celebrate by making a yummy fruit smoothie - although I'm really afraid of this mixer...
(https://t4.ftcdn.net/jpg/06/00/43/27/240_F_600432797_iC6gVj94tCdTjaJ6rYnhxt8g4LO3GBrx.jpg) (https://stock.adobe.com/images/mixer-of-fruit-juice-on-the-wooden-table-ai-generative/600432797)
The mixer is clearly Borg and in the process of assimilating the kitchen.
Resistance is futile!
R = V/I
-
One month after the OP, the Adobe Stock genAI collection (tagged; there are more that are not) is over 12 million - 12,004,534
We can celebrate by making a yummy fruit smoothie - although I'm really afraid of this mixer...
(https://t4.ftcdn.net/jpg/06/00/43/27/240_F_600432797_iC6gVj94tCdTjaJ6rYnhxt8g4LO3GBrx.jpg) (https://stock.adobe.com/images/mixer-of-fruit-juice-on-the-wooden-table-ai-generative/600432797)
The mixer is clearly Borg and in the process of assimilating the kitchen.
Resistance is futile!
or a hand crank?
-
The genAI collection is now 13,051,798 - world domination lies ahead!
I forgot to celebrate the 13 million milestone, so I thought I'd grab something to eat and then go on a trip.
I grew up in the UK and a Cornish Pasty sounded lovely - until they brought this concoction with red peppers and other foreign objects. I sent it back
(https://t3.ftcdn.net/jpg/06/03/84/58/240_F_603845838_B4L3AclXj6Hnt2ije9zQ97PnoSE2yRgV.jpg) (https://stock.adobe.com/images/close-up-of-a-cornish-pastry-british-bakery-from-cornwall/603845838)
Then I thought I'd try a different cuisine and ordered Aguachile, but what came out was an onion blanket with nothing much underneath, so I sent that back too.
(https://t3.ftcdn.net/jpg/06/25/88/44/240_F_625884450_jiJVDVr0LwwEaDCHDaeYiiyn8w1uwgzT.jpg) (https://stock.adobe.com/images/aguachile-showcasing-the-vibrant-color-of-the-shrimp-marinated-in-chili-and-lime-garnished-with-fresh-cucumbers-and-red-onions-shot-from-above/625884450)
I didn't want to miss my flight, so I borrowed a yellow skateboard to hurry over, but when I looked more closely, it was a strange seated model I thought I might injure myself with...
(https://t4.ftcdn.net/jpg/06/25/89/01/240_F_625890199_IZquJmZxIQ9scl7sRWDVHtlLyrYRXjW2.jpg) (https://stock.adobe.com/images/yellow-skateboard-or-skating-surf-board-on-vibrant/625890199)
... so I called a taxi. Looked dodgy, but I was out of options!
(https://t3.ftcdn.net/jpg/06/13/73/96/240_F_613739677_KeJGdJhXS4r8u2Z056c7f05FQ2pViBmy.jpg) (https://stock.adobe.com/images/yellow-rusty-old-taxi-desktop-wallpaper-in-the-smoke-high-resulation-uhd-4k/613739677)
Finally at the airport. Thinking my troubles were over, I looked up at the stairs ...
(https://t4.ftcdn.net/jpg/06/16/14/75/240_F_616147567_cljobhUrkMV5PiZxt3zt7AHaRvwYMubO.jpg) (https://stock.adobe.com/images/boarding-commercial-airplane-with-red-carpet-presentation-generative-ai/616147567)
I finally sat down, starving hungry by this time, and then my heart sank as I looked at the food and drink...
(https://t4.ftcdn.net/jpg/05/74/54/39/240_F_574543967_9lwHuaeapDFRwKZ0VswTEjUqjZKQxgVo.jpg) (https://stock.adobe.com/images/interior-of-luxurious-private-jet-with-leather-seats-generative-ai/574543967)
-
Well, there are certainly double standards in reviewing.
Good and useful images that get rejected for "quality issues" (yes, looking at you Adobe) while AI seems to get a free pass-through, no matter how much they're off.
And I get it, AI generated content that looks amazing at first sight (and awful with closer inspection) is easy to produce, and agencies must be flooded with that.
I also get that agencies don't want to miss out on the AI train.
But they have to realize that sloppy reviewing floods their database with junk, and in the end, this doesn't do any good for their customers either.
In the end, they pass the reviewing over to their customers.: bad content doesn't sell and gets buried by the search engine algorithm.
That system works fine for the occasional error that reviewers make, but if you flood your database with junk, nobody will bother anymore to wade through it.
We have the "junk" issue with photography too, but not in this volume, not at this pace.
-
You know what. I take back everything I said about embracing AI. If this is the quality that gets accepted by Adobe; practically useless images, inaccurate depictions of people, locations and food that no client would want to buy... I'm afraid buyers will leave the platform and this AI experiment will backfire. It's already flooding the database and burying the real stuff, so buyers will have to search for needles in an artificial haystack.
-
However bad you think the accepted genAI images are, go take a look at a few real world types of images - things you think a buyer might be searching for - and you'll realize it's worse.
These examples are just a selection from page 1 of the default (Relevance) search. I'm ignoring the fantasy images (even the cute mouse with sunglasses and a bow tie) as that's not where the issue is.
Somehow this woman made it on a bike with one pedal - possibly because she has part of one leg missing - and strange mangled spokes. Extra points for her not holding the handlebars, plus bad Photoshop work between the handlebar and brake lever. Cheating to cover the suspicious construction of the seat to frame connection with her dress!
(https://t4.ftcdn.net/jpg/06/13/26/95/240_F_613269571_PqSxqHWGg8h8Qh58L4djzMixvQ8ukmN0.jpg) (https://stock.adobe.com/images/woman-from-behind-in-white-dress-and-with-a-bicycle-generative-ai/613269571)
Apparently Byzantine painters were no better with correct number and shapes of hands and feet than AI - who knew?
(https://t3.ftcdn.net/jpg/06/13/24/10/240_F_613241028_Ap5KtwWTy1ChjoCCIzwsuyan1FRPxSnf.jpg) (https://stock.adobe.com/images/biblical-representation-of-jesus-*-in-byzantine-art-ai-generativ/613241028)
This is keyworded as Santorini - as are 4,037 other genAI images. Did not count how many of those look like photos - some are labeled as watercolor or other artistic style. The pseudo paintings or pseudo illustrations seem less problematic in that they are clearly not purporting to show real life.
(https://t3.ftcdn.net/jpg/06/19/50/26/240_F_619502666_dXpoBqanfdxsIhwPIt9nk4uQgV96uAFN.jpg) (https://stock.adobe.com/images/church-in-island/619502666)
"Image of long rows of green beds with growing cabbage or lettuce in a large farmer's field". I'd bet on lettuce - though I'm not a farm or vegetable expert.
(https://t4.ftcdn.net/jpg/06/14/18/39/240_F_614183930_aXeAdkHL34HvE9FMQ0Ht3dVIf8imrS15.jpg) (https://stock.adobe.com/images/image-of-long-rows-of-green-beds-with-growing-cabbage-or-lettuce-in-a-large-farmer-s-field-generative-ai/614183930)
Here's what a row of cabbages (https://stock.adobe.com/images/rows-of-ripe-cabbage-under-the-evening-sky/168702280) actually looks like
"A happy couple stands smiling in the driveway of a large house with solar panels installed". If you look at their front door (left) their smiles must result from having figured how to get out and down the steps without breaking their necks!
(https://t3.ftcdn.net/jpg/06/10/53/40/240_F_610534082_VxcBh43iCoPiWabWEm4J9q4MBr8jydYl.jpg) (https://stock.adobe.com/images/a-happy-couple-stands-smiling-in-the-driveway-of-a-large-house-with-solar-panels-installed-generative-ai/610534082)
Chair and stool legs, along with hanging lights, cabinet handles and other details, seem to confuddle AI - don't sit on chairs in AI-world.
(https://t4.ftcdn.net/jpg/06/10/76/37/240_F_610763739_FvqeFMTnvVBhjHSjmEwB2FM39oxJQoxC.jpg) (https://stock.adobe.com/images/modern-domestic-kitchen-design-with-luxury-wood-material-and-stainless-steel-generated-by-ai/610763739)
Impossibly hip and stylish high school student can't escape AI's problems with digits (mangled pinky)
(https://t3.ftcdn.net/jpg/06/16/48/22/240_F_616482248_xB7q8eM3atR4Huk1ZSSz3BWiqdvwNd9S.jpg) (https://stock.adobe.com/images/high-school-student-engaging-in-discussion-during-class-lecture/616482248)
It's all really depressing - instead of celebrating the strengths of genAI images, we're getting cheap and wonky knock-offs of real world places and things. I'm hoping the market (buyers; Adobe clearly loves welcoming this stuff into the collection) speaks and ignores things that aren't useful. Or which get them articles in the Guardian (https://www.microstockgroup.com/fotolia-com/adobe-stock-needs-a-visible-label-on-genai-images-like-editorial-and-premium/msg590145/#msg590145) :)
-
Example here (Blurred hatchet job is me to hide my ID) but these are Firefly generated marine life and the prompt use.
Seahorse with its tail connected to its neck?
Dolphin thats some sort of Orca?
Reef shark thats a Whale shark with odd fins?
1 legged octopus.
and so on
Dont think i'll submit this somehow.
-
I have a large folder of Firefly experiments. It's still in beta. I haven't produced a single usable image (for the things I was interested in working on), and the workflow seems awkward to me -not sure how I would use it if it was out of beta.
Earlier in the week I tried the generative fill option on an image I'd created with firefly and it was funny - not intentionally. I wanted to replace the floor surface in an interior shot. It couldn't automatically select the floor, or even automatically extend my manual selection. The perspective and scale on the fills wasn't right. A mess.
-
I've tried Firefly beta too for some project, but the results were clearly even less accurate than Midjourney.
It does work OK for small fills/extensions of floors, walls or skies; the trick is to also select the edge of the pattern you want to extend, so Firefly has some information and space to work with. Then Firefly can blend it nicely with the existing textures.
But for new objects (I tried palm trees) it comes up with ugly deformed trees which didn't fit at all in my composition. Either Firefly is still too primitive or finding the right prompts is truly an art form.
-
It's all really depressing - instead of celebrating the strengths of genAI images, we're getting cheap and wonky knock-offs of real world places and things. I'm hoping the market (buyers; Adobe clearly loves welcoming this stuff into the collection) speaks and ignores things that aren't useful. Or which get them articles in the Guardian (https://www.microstockgroup.com/fotolia-com/adobe-stock-needs-a-visible-label-on-genai-images-like-editorial-and-premium/msg590145/#msg590145) :)
I think that's the real question: how will we all, as customers, accept these images that are meant "to illustrate" a product or a service?
The burger advertised by McDonalds doesn't look like the burger you get at their counters.
The car that I bought, and was advertised as a shiny car driving to smooth city traffic against flashy buildings, while reality is more like getting stuck in shabby neighborhoods.
Santorini never looks like it get's advertised in travel brochures. People everywhere!
And I have never seen a couple smiling on their driveway because they just installed solar panels.
The point is: we, as a public, already accept an idealized version of reality, and we know that the real world looks more bleak.
We might as well getting pushed in accepting the next version of it. Impressions of how it could look like, with shabby and wonky AI imagery.
After all, on every advertisement there's a small notice: real product might differ from advertised product.
-
The car that I bought, and was advertised as a shiny car driving to smooth city traffic against flashy buildings, while reality is more like getting stuck in shabby neighborhoods.
...
After all, on every advertisement there's a small notice: real product might differ from advertised product.
Most people understand that they buy just the car, not the neigbourhood.
However, if the picture of the car they order looks like a Porsche and what they get looks more like a Volkswagen, they might get cranky.
-
It's all really depressing - instead of celebrating the strengths of genAI images, we're getting cheap and wonky knock-offs of real world places and things. I'm hoping the market (buyers; Adobe clearly loves welcoming this stuff into the collection) speaks and ignores things that aren't useful. Or which get them articles in the Guardian (https://www.microstockgroup.com/fotolia-com/adobe-stock-needs-a-visible-label-on-genai-images-like-editorial-and-premium/msg590145/#msg590145) :)
I think that's the real question: how will we all, as customers, accept these images that are meant "to illustrate" a product or a service?
The burger advertised by McDonalds doesn't look like the burger you get at their counters. ....
After all, on every advertisement there's a small notice: real product might differ from advertised product.
Point taken - I've been licensing stock and looking at ads/marketing materials for decades. "Authentic" in stock isn't real life - and if you look at all the "shiny-happy people" pictures, genAI or old-school, they're better lit, tidier, posed and typically younger/prettier than most people's everyday life.
However, there's a huge difference between cleaning up, polishing and giving a sunset glow to real life and making things up that sort-of look like the real thing.
Here's an example. Below is a genAI image of Sheffield town hall, Yorkshire, UK. This link will, courtesy of google (https://www.google.com/maps/place/Sheffield+Town+Hall/@53.3802394,-1.4687796,3a,75y,90t/data=!3m8!1e2!3m6!1sAF1QipPNN-Qq7zldVu5rNWe3mOL09rEbBDH1Pk-wyzJT!2e10!3e12!6shttps:%2F%2Flh5.googleusercontent.com%2Fp%2FAF1QipPNN-Qq7zldVu5rNWe3mOL09rEbBDH1Pk-wyzJT%3Dw114-h86-k-no!7i4080!8i3072!4m7!3m6!1s0x48798281f4658e11:0xd1f77f34d3f67305!8m2!3d53.3802394!4d-1.4687796!10e5!16zL20vMDF2Yzdt?entry=ttu), show you a large set of pictures of the town hall you'd see if you went to Sheffield.
(https://t3.ftcdn.net/jpg/05/85/82/22/240_F_585822236_wDCkPItcznrslzq2J7R8ffhVGNGYj1Sl.jpg) (https://stock.adobe.com/images/view-of-sheffield-city-council-and-sheffield-town-hall-in-autumn-generative-ai/585822236)
It's not just the autumn trees or the warm color tone. The genAI image has the clock tower in a different place with respect to the other buildings; the windows and floors are different from the real building. Just about every architectural detail in the genAI building is wrong, although it has a similar feel overall.
I have many hundreds of examples from Adobe Stock of this sort of problem - the town hall image isn't an exception. There are images from cities all over the world; of objects that are impossible or defy the laws of physics (but which are presented as real life not fantasy); human or animal anatomy, vegetable/fruit structure that are wrong (not ironic or artistic or fantastic; just wrong); flags or other emblems improperly rendered.
As long as this type of work is clearly labeled as genAI, buyers can choose. IMO Adobe would be well served by removing the obvious "oops" images (https://stock.adobe.com/images/stylish-scandinavian-composition-of-living-room-with-design-armchair-mock-up-poster-frame-personal-accessories-in-modern-home-decor-generative-ai/570201793). Although perhaps the aircraft carrier with wings (https://stock.adobe.com/images/a-military-aircraft-carrier-with-fighter-jets-generative-ai/574947471) should stay :)
-
...
Santorini never looks like it get's advertised in travel brochures. People everywhere!
And I have never seen a couple smiling on their driveway because they just installed solar panels.
The point is: we, as a public, already accept an idealized version of reality, and we know that the real world looks more bleak. ...
actually i have many images of Santorini and other tourist areas with no people - sometimes a few people can be replaced, but traveling off season can give those human-free pix
the bigger question is why flood an already flooded market? for my AI i'm concentrating on pix that are difficult or impossible to get otherwise.
-
Another general topic for contributors to contemplate - genAI images can end up looking very much alike.
I was looking at a set of images in today's new approvals and thinking there was something wrong with the sort order - that I was seeing old images from a couple of weeks ago.
Turns out there are two separate portfolios each with a batch of very, very similar sunset-beach-palm-tree pictures. No copying, just whatever tool they used has apparently only got so many ways to deliver results.
Older image batch (https://stock.adobe.com/search?creator_id=211459262&filters%5Bcontent_type%3Aphoto%5D=1&filters%5Bcontent_type%3Aillustration%5D=1&filters%5Bcontent_type%3Azip_vector%5D=1&filters%5Bcontent_type%3Avideo%5D=1&filters%5Bcontent_type%3Atemplate%5D=1&filters%5Bcontent_type%3A3d%5D=1&filters%5Bfetch_excluded_assets%5D=1&filters%5Bcontent_type%3Aimage%5D=1&order=relevance&safe_search=1&k=sunset+beach+shore&limit=100&search_page=1&search_type=pagination&acp=&aco=sunset+beach+shore&get_facets=0) - Today's image batch (https://stock.adobe.com/search?creator_id=211578057&filters%5Bcontent_type%3Aphoto%5D=1&filters%5Bcontent_type%3Aillustration%5D=1&filters%5Bcontent_type%3Azip_vector%5D=1&filters%5Bcontent_type%3Avideo%5D=1&filters%5Bcontent_type%3Atemplate%5D=1&filters%5Bcontent_type%3A3d%5D=1&filters%5Bfetch_excluded_assets%5D=1&filters%5Bcontent_type%3Aimage%5D=1&order=relevance&safe_search=1&k=sunset+beach&limit=100&search_page=1&search_type=usertyped&acp=&aco=sunset+beach&get_facets=0)
I'm not trying to argue that only one contributor can have sunset-beach-palm-tree images, but do think it's worth considering what value to the buyer there is in having such nearly-cloned images. I also wouldn't argue with the composition of these images - they're all very pretty. And there are none of the heffalump traps that genAI so often falls into in the subject matter.
If you imagine a number of photographers going to a particular beach to shoot a pretty sunset on different days at different times of the year - or even the same photographer going back multiple times - I can't imagine them producing such homogenous results as these two sets of images. Our planet's weather systems and land-ocean interfaces are highly variable - it would be very hard to get the same result twice if that was what you wanted.
Does this mean that with the current data set training, at some point we'll be getting more repeats (for similar prompts)? I know nothing about these two contributors, so it is possible that this is some sort of collaboration and this is all really one big set of images that was divvied up and uploaded separately.
-
Stock photography is coming to an end has been the most consistent topic on this forum for the last 15 years. And yet, here we are.
15 years ...??? ;D
-
Jo Ann Snover - Great post ... ;D
-
This is crazy. >:(
-
Turns out there are two separate portfolios each with a batch of very, very similar sunset-beach-palm-tree pictures. No copying, just whatever tool they used has apparently only got so many ways to deliver results.
My guess is that these two accounts belong to the same contributor. Probably using multiple accounts to get around upload limits.
-
The similar issue between different ports, actually duplicates, is going is going to become huge.
People just sort by downloads and then copy the bestsellers.
Also people keep copying prompts they find somewhere, or even take the description of some images they like, add it to an ai and voila…they basically get the same picture and upload that.
I used to write detailed image descriptions. Now I make them much shorter and make sure to leave out some essential elements.
But ai makes copying really, really easy.
Most ai producers use midjourney, so you keep coming across ports that all look exactly alike, as if midjourney is the producer and the content is just being distributed through different ports.
One of the reasons I am not using midjourney…but even with other engines you see the duplicates.
So will always keep taking photos. Yes, there are copy cats, but it is never exactly the same.
-
But they have to realize that sloppy reviewing floods their database with junk, and in the end, this doesn't do any good for their customers either.
In the end, they pass the reviewing over to their customers.: bad content doesn't sell and gets buried by the search engine algorithm.
That system works fine for the occasional error that reviewers make, but if you flood your database with junk, nobody will bother anymore to wade through it.
I remember some time ago in the past when reviewers at AS were really strict and fussy and it was hard to get some content accepted. In more recent times, they seem to be getting less strict and it's not often when I get rejections these days.
-
Here's an article with accompanying images showing an AI generated model that's 'active' on social media. Looks like 'she' has a lot of followers despite not being real.
https://au.lifestyle.yahoo.com/stunning-bikini-model-racks-up-thousands-of-fans-online-despite-not-being-real-015746428.html
-
The collection is now over 13.5 million. Just a handful of the recent crazy acceptances...
Even if I were taller, I don't know how I'd get into this chair...
(https://t3.ftcdn.net/jpg/06/24/93/20/240_F_624932015_mR10TJLYlCZS5hnluOblMyomf6TMW4kY.jpg) (https://stock.adobe.com/images/beautiful-armchair-red-chair-office-chair-design-images-generated-by-ai-technology/624932015)
The basics of a bicycle really aren't that complicated
(https://t4.ftcdn.net/jpg/06/26/16/11/240_F_626161149_MCyZLUfmvkRsuUNlfysB9jeyfA8ru2y0.jpg) (https://stock.adobe.com/images/a-vintage-bicycle-leaning-against-sea-at-beach-in-morning-wicker-basket-with-artificial-flowers-on-the-bike/626161149)
Customer service: Why are we getting all these complaints about newly-arrived goods being soaking wet?
(https://t3.ftcdn.net/jpg/06/28/55/22/240_F_628552292_ZIjlZfTAXOgiM0PakTeP3AM4DxEe5gsh.jpg) (https://stock.adobe.com/images/a-row-of-cargo-containers-ready-for-international-export-with-a-busy-shipping-pier-as-the-backdrop-embodying-the-seamless-logistics-and-extensive-reach-of-containerized-shipping-generative-ai/628552292)
How dare you start commenting on my extra legs!
(https://t3.ftcdn.net/jpg/06/09/92/66/240_F_609926602_J46z46qgXNnXIgPfU7RgAJXgUNq2hJ1T.jpg) (https://stock.adobe.com/images/pool-in-hotel/609926602)
This clearly isn't cheesecake, but even as pasta I wouldn't put that in my mouth!
(https://t4.ftcdn.net/jpg/06/27/80/47/240_F_627804768_1UiJNLHZWj0One9k1JUTKHcB9V0lYxzD.jpg) (https://stock.adobe.com/images/cheesecake-with-berries-and-straberry-and-cream-chocolate-mix-ai-generative/627804768)
This is closer, but also doesn't look like "cheesecake with berries and straberry"
(https://t3.ftcdn.net/jpg/06/27/80/46/240_F_627804664_sv7dL5xkUKAglt3PPJYCEIUmxbw95RwN.jpg) (https://stock.adobe.com/images/cheesecake-with-berries-and-straberry-and-cream-chocolate-mix-ai-generative/627804664)
It's all completely mad!
-
Please keep this up - its my morning dose of humour!
-
The month long waiting time for reviews is getting aggravating.
-
Just two weeks after crossing the 13 million mark, the genAI collection is now over 14 million - 14,062,823
-
Just two weeks after crossing the 13 million mark, the genAI collection is now over 14 million - 14,062,823
Amazing! What would be interesting to know is for those selling AI images as stock, how do the $$$ work out? In other words what is the net revenue selling genAI stock vs. the cost of "membership" to the genAI websites?
-
Just two weeks after crossing the 13 million mark, the genAI collection is now over 14 million - 14,062,823
Amazing! What would be interesting to know is for those selling AI images as stock, how do the $$$ work out? In other words what is the net revenue selling genAI stock vs. the cost of "membership" to the genAI websites?
On Midjourney it's 10$ (minimum) vs whatever you get from Adobe Stock and Dreamstime:
https://www.microstockgroup.com/37744/37744/msg590316/#msg590316 (https://www.microstockgroup.com/37744/37744/msg590316/#msg590316)
-
AI optimists say that genAI images will keep getting better, but I'm not seeing much evidence of that - although genAI at Adobe Stock is light years ahead of SS's sad AI collection
Better with hands and fingers?
(https://t4.ftcdn.net/jpg/06/27/24/67/240_F_627246781_6KK9ivkoJpk4JEZ8gkKwS8BMsn19Wvul.jpg) (https://stock.adobe.com/images/female-hand-with-blue-and-white-pills-on-wooden-table/627246781)
Better with staircases?
(https://t3.ftcdn.net/jpg/06/33/20/80/240_F_633208006_595HyjDuHHYlnaayBb0q1NQQCi783LZt.jpg) (https://stock.adobe.com/images/living-room-loft-in-industrial-style-generative-ai/633208006)
Better at important details of the real world? Such as supermarket aisles where one person can barely fit through, forget carrying a basket or pushing a cart...
(https://t3.ftcdn.net/jpg/06/19/63/68/240_F_619636898_r9msepUzpApdra5CREvzbP7gWCYptfGp.jpg) (https://stock.adobe.com/images/woman-comparing-products-in-a-grocery-store-considering-nutrition-prices-and-ingredients-demonstrating-informed-consumer-behavior-generative-ai/619636898)
Better at obeying the laws of physics? And this is part of a series of impossible turtle behavior. (https://stock.adobe.com/images/turtle-on-the-rocks/629310456)
(https://t3.ftcdn.net/jpg/06/29/31/04/240_F_629310401_LCF1qsziqx6d2JMw5ZfEerKXZpXp35sI.jpg) (https://stock.adobe.com/images/turtle-on-the-beach/629310401)
Enough of the negativity! Halloween's not far away and there's a mass of genAI content of pumpkins and orange flowers decorating porches, patios and steps. Many of them seem to have forgotten that doors open and people have to walk down steps. Those pumpkins are disturbing...
(https://t3.ftcdn.net/jpg/06/26/69/94/240_F_626699424_29NtUEhOVvj2DYXvPULQmltDsY2DyTEW.jpg) (https://stock.adobe.com/images/halloween-pumpkins-on-steps-of-front-porch-exterior-home-decorations-seasonal-decor/626699424)
I'll spare you the creativity gone wrong - ramen burgers, vintage clothespins and spooky coaches!
-
Every time I see some bizarre details and think..oh I will just leave that, nobody will notice...then I think: but wait!
What if Jo Ann finds it??
You are making me a better ai artist...:)
-
Every time I see some bizarre details and think..oh I will just leave that, nobody will notice...then I think: but wait!
What if Jo Ann finds it??
You are making me a better ai artist...:)
;D ;D
I have the same fear... I'm afraid Jo Ann will scold me ;D ;D
-
Just two weeks after crossing the 13 million mark, the genAI collection is now over 14 million - 14,062,823
Amazing! What would be interesting to know is for those selling AI images as stock, how do the $$$ work out? In other words what is the net revenue selling genAI stock vs. the cost of "membership" to the genAI websites?
On Midjourney it's 10$ (minimum) vs whatever you get from Adobe Stock and Dreamstime:
https://www.microstockgroup.com/37744/37744/msg590316/#msg590316 (https://www.microstockgroup.com/37744/37744/msg590316/#msg590316)
The plan pricing is confusing things like "3 jobs, 10 jobs waiting in queue". How does that translate into things like: "For $10 month, how many images can the user generate?"
-
Every time I see some bizarre details and think..oh I will just leave that, nobody will notice...then I think: but wait!
What if Jo Ann finds it??
You are making me a better ai artist...:)
I live to serve!! :)
I've been editing images for decades and spent oodles of time removing sensor spots, logos, numbers on sails, anything that could look like a number on a sail to an inspector, and so on. I have lots of scar tissue from lots of rejections but I learned what a clean stock image should look like. (Obviously you also need to have a decent composition to start with).
I'm mostly furious that the agencies have just abandoned so many of their standards and are accepting stuff customers can't possibly use. I'd love to scold the inspectors but I can't :)
Sorry if it comes off the wrong way.
-
Just two weeks after crossing the 13 million mark, the genAI collection is now over 14 million - 14,062,823
Amazing! What would be interesting to know is for those selling AI images as stock, how do the $$$ work out? In other words what is the net revenue selling genAI stock vs. the cost of "membership" to the genAI websites?
On Midjourney it's 10$ (minimum) vs whatever you get from Adobe Stock and Dreamstime:
https://www.microstockgroup.com/37744/37744/msg590316/#msg590316 (https://www.microstockgroup.com/37744/37744/msg590316/#msg590316)
The plan pricing is confusing things like "3 jobs, 10 jobs waiting in queue". How does that translate into things like: "For $10 month, how many images can the user generate?"
You can generate few hundreds, but many of these won't be good enough for licensing.
-
Every time I see some bizarre details and think..oh I will just leave that, nobody will notice...then I think: but wait!
What if Jo Ann finds it??
You are making me a better ai artist...:)
I live to serve!! :)
I've been editing images for decades and spent oodles of time removing sensor spots, logos, numbers on sails, anything that could look like a number on a sail to an inspector, and so on. I have lots of scar tissue from lots of rejections but I learned what a clean stock image should look like. (Obviously you also need to have a decent composition to start with).
I'm mostly furious that the agencies have just abandoned so many of their standards and are accepting stuff customers can't possibly use. I'd love to scold the inspectors but I can't :)
Sorry if it comes off the wrong way.
Do we know for a fact that it's humans and not algorithms approving/rejecting images? I'm sure some "quality" rejections are automated but it wouldn't shock me if they tried to automate the rest of it too.
-
From what they take or decline I would think this was a human decision. Doesn't look automated to me.
-
Just two weeks after crossing the 13 million mark, the genAI collection is now over 14 million - 14,062,823
Yup, it seems that there's over 100k approved images that are AI-gen per day. That's.. a lot.
-
If you look at the stats below the forum we now often have over 2000 people online reading, although often just 20 active, registered users.
It used to be a few hundred, or maybe 500.
I think this is all due to ai. Millions of people trying to learn how to make and sell ai images.
Even if you just have a smartphone, you can be a great "photographer".
Can't blame them for trying.
They will soon realise that working with stock agencies is not as simple as the youtubers make it sound.
-
If you look at the stats below the forum we now often have over 2000 people online reading, although often just 20 active, registered users.
It used to be a few hundred, or maybe 500.
I think this is all due to ai. Millions of people trying to learn how to make and sell ai images.
Even if you just have a smartphone, you can be a great "photographer".
Can't blame them for trying.
They will soon realise that working with stock agencies is not as simple as the youtubers make it sound.
Interesting, never noticed the numbers...
-
Just two weeks after crossing the 13 million mark, the genAI collection is now over 14 million - 14,062,823
Amazing! What would be interesting to know is for those selling AI images as stock, how do the $$$ work out? In other words what is the net revenue selling genAI stock vs. the cost of "membership" to the genAI websites?
The content sells like any other content.
If you create something from a needed niche with the help of ai it sells just as great as if you had done a regular shoot.
If you upload stuff noone needs, you have no sales.
I have 637 accepted ai files on Adobe and 129 downloads. Made 114.83 USD so far with an rpd of .89. Most files have only sold once and only 1 file has more than 5 sales. None have 10. So no real bestsellers (yet).
But I have paid over 2000 dollars in credits for ai use (not using midjourney). Expect to reach 3k by end of the year. Includes 600 dollars from the last 4 years when I was just prompting for fun.
Next year I should spend a lot less, because my prompting is a lot better and because the prices to use ai commercially are coming down. Ideally I will work only with firefly, if the quality is good enough and if they offer producers a good deal (or maybe free use, with enough uploads).
It took me a lot of time to learn my way around ai. And every time I attempt a new series, I need a new set of prompts.
Prompts also "break down" frequently if I just change something simple, like pink instead of red or morning light instead of evening light. Suddenly I get unusable pixel mash.
I also post process my files a lot in photoshop, because so many details are broken when you zoom into 100%.
However…I think having ai content brings a lot of eyeballs to the port. I am certainly seeing a strong uptick in overall sales.
Midjourney probably gets you the best content with little post processing needs. Downside…everyone is using midjourney and you run the risk of having just another "metoo" port.
I try to minimize that by checking the interesting files after prompting. If I see 10 other near identical uploads, I discard mine.
Hoping to have 1200 files from different genres by end of the year. But I am already slowing down the prompting and doing more normal photography and especially more video.
Because the other big downside is that at the moment only Adobe has a real ai collection, so I don't have enough content for the other agencies.
This is my experience.
The youtubers are all screaming how easy it is to make thousands of dollars a month with ai.
So either I am doing the wrong content or the wrong genres or…strangers on the internet are not telling the truth…
-
Thanks! very interesting to hear actual real world experience.
-
We all benefit from sharing real information.
Also working with ai has taken away the fear that ai will replace me. Maybe in 10-20 years, but not now. It is a tool I can learn to use. Nothing more, nothing less.
-
Made 114.83 USD so far
But I have paid over 2000 dollars in credits for ai use
If you are spending so much more than what you are earning that makes you a example how not to use AI for microstock. :o
I've only spent $71.40 on AI and earned back a multiple of it. If I hadn't, I would stop spending money on AI.
-
Made 114.83 USD so far
But I have paid over 2000 dollars in credits for ai use
If you are spending so much more than what you are earning that makes you a bad example how not to use AI for microstock. :o
I've only spent $71.40 on AI and earned back a multiple of it. If I hadn't, I would stop spending money on AI.
My son spent the same exactly (71.4$) and earned 284$ in the last 5 months.
-
He uploaded a lot more files, didn‘t he? 2000? I have 630.
Plus a lot of my content is seasonal, for christmas and now starting with easter, so the sales will come later. But for a good harvest you have to upload at least 6 months in advance.
And with the Adobe delay, maybe 9 months early. Hence…easter now, not October.
Everyone who is using midjourney doesn‘t have my cost basis.
I am not recommending to do it my way.
But I am not planning to use midjourney and I will not upload too many files in one go.
I am building my port very slowly with intention.
But for sales it is exactly like I said. Upload needed content, you get sales.
Upload stuff just for fun, no sales.
Don‘t worry, I will get my money back.
eta
unless you pay for the private room everything on midjourney is public. And we have an army of aggressive copycats on our heels.
and with ai files you will not be able to file a dmca takedown complaint. copycats are free to upload the exact same image, or with subtle variation without any consequences.
Everyone has to make their own decisions how to deal with that.
I have several thousand files good enough for processing and uploading. But I never upload a complete series in one go.
Not with photos or video or ai.
I would rather have a small port that is efficient.
But that is a personal choice.
-
He uploaded a lot more files, didn‘t he? 2000? I have 630.
Plus a lot of my content is seasonal, for christmas and now starting with easter, so the sales will come later. But for a good harvest you have to upload at least 6 months in advance.
And with the Adobe delay, maybe 9 months early. Hence…easter now, not October.
Everyone who is using midjourney doesn‘t have my cost basis.
I am not recommending to do it my way.
But I am not planning to use midjourney and I will not upload too many files in one go.
I am building my port very slowly with intention.
But for sales it is exactly like I said. Upload needed content, you get sales.
Upload stuff just for fun, no sales.
Don‘t worry, I will get my money back.
eta
unless you pay for the private room everything on midjourney is public. And we have an army of aggressive copycats on our heels.
and with ai files you will not be able to file a dmca takedown complaint. copycats are free to upload the exact same image, or with subtle variation without any consequences.
Everyone has to make their own decisions how to deal with that.
I have several thousand files good enough for processing and uploading. But I never upload a complete series in one go.
Not with photos or video or ai.
I would rather have a small port that is efficient.
But that is a personal choice.
The big problem is that millions of images are added monthly to the Ai collection, so quantity may serve you better. You may create some great quality images that got lost in the huge Ai collection. So, I'm afraid that time goes against your strategy.
As I said before, I always warn my son that this domain is changing so fast and he should not put big hopes on its future.
-
I have been doing this for nearly 20 years. Agencies take 1-2 million PHOTOS every week. With your logic nobody would have any sales uploading photos. (Shutterstock used to publish weekly stats of how much they accept every week. They stopped publishing when we went into an upload strike, because you could see how the uploads have dropped)
I only have 2400 files in my port and have daily sales on Adobe. And took a ten year break of basically not uploading.
Stock is not a volume game. It is a USEFUL game.
Upload something useful and you will have sales, because you are only competing with the files in the niche.
Ice cream does not compete with spaghetti, empty rooms not with people.
100k ai files a week is very little, because it is spread over a very wide range of genres.
Adobe probably gets more uploads of animals or sunsets or flowers than ai in a week.
And how much of the uploaded ai content is really saleable and had trend research?
Mostly it is metoo material.
So, I will stick to my plan and enjoy my results :)
But of course, if someone wants to upload 1000 files a week, please do. But if you want sales, you have to make them useful for customers.
If I could generate 1000 files every week that are NOT similar to each other and come from completely different themes and need no post processing I would upload them.
But I cannot generate that many files from completely different themes so quickly because it takes a lot of time to analyze trends and what is already in the collections.
ETA2
simple example - how many easter themed ai images are being uploaded RIGHT NOW. Because that cohort of files will be my competitors for the season.
Plus of course any other easter photos/illustrations being uploaded now.
Your files move in certain time batches. For optimal harvest you need very good quality, a useful, slightly different twist from what is already available and ...good timing. (for seasonal material)
You can of course prompt endless happy business people that sell all year round. But I would have to research that genre before I start prompting.
-
unless you pay for the private room everything on midjourney is public.
Isn´t that the same with Dall-E?
And with Stable Diffusion (unless you run it on your own computer)?
-
Don‘t worry, I will get my money back.
Yeah, but others, including me, have already gotten their money back AND earned a multiple of it. So my point stands. You way does not seem very efficient.
unless you pay for the private room everything on midjourney is public. And we have an army of aggressive copycats on our heels.
and with ai files you will not be able to file a dmca takedown complaint. copycats are free to upload the exact same image, or with subtle variation without any consequences.
Yea, but what does it matter? Public, not public. I have still earned much more money with AI than you and only spent a fraction.
Will I need to be able to file a dmca takedown complaint? Why? Why should I care. I have zero personal attachment to these images, I spent zero effort on them (expect for the keywords, which is the one part that is not public on Midjourney) and I hardly spent any money on them. A fraction of a cent per image a best. I really do not care. Unlike with my real actual photos, where I spend countless of hours of work and also money on. That's where I care. That's where I do not want my work to be stolen (which AI has basically done!!!!). I did not create these AI images! An AI did! I don't even feel like they should belong to me.
Microstock is about earning money. And if, at the bottom line, no matter whether "you will get back your money" or "images are public" or whatever else argument you try to bring forth, someone else is earning much more money than you, it all does not matter - The method of the person who earns more money is better than yours.
The ratio of how much you have spent on AI and what little you have earned back is insane! And you constantly use your own experience to explain how AI will not make real photographers pointless, how it will not bring down the microstock business and how not everyone can do it. How it was hard and took effort and money. When all of this is simply not true. Your method is simply not efficient.
You do not understand how EASY and CHEAP and FAST everyone could now make money with AI images.
Because you have not figured out for yourself how it works yet.
And once everyone finally figures it out no one will be able to earn any decent money anymore, because the pie is not big enough. Expect the companys who bulit the AIs. The only ones who will in the long run profit from this.
-
He uploaded a lot more files, didn‘t he? 2000? I have 630.
Plus a lot of my content is seasonal, for christmas and now starting with easter, so the sales will come later. But for a good harvest you have to upload at least 6 months in advance.
And with the Adobe delay, maybe 9 months early. Hence…easter now, not October.
Everyone who is using midjourney doesn‘t have my cost basis.
I am not recommending to do it my way.
But I am not planning to use midjourney and I will not upload too many files in one go.
I am building my port very slowly with intention.
But for sales it is exactly like I said. Upload needed content, you get sales.
Upload stuff just for fun, no sales.
Don‘t worry, I will get my money back.
eta
unless you pay for the private room everything on midjourney is public. And we have an army of aggressive copycats on our heels.
and with ai files you will not be able to file a dmca takedown complaint. copycats are free to upload the exact same image, or with subtle variation without any consequences.
Everyone has to make their own decisions how to deal with that.
I have several thousand files good enough for processing and uploading. But I never upload a complete series in one go.
Not with photos or video or ai.
I would rather have a small port that is efficient.
But that is a personal choice.
If I'm looking at the correct portfolio, you have a lot of fantastical landscapes, steampunk, watercolor genAI stuff etc.
I get it, you're targeting a niche, but it may not be the most cost-effective use of your time.
I'm using SDXL on ComfyUI (previously SD 1.5 with various models), and spent $0 on producing any genAI images. Just electricity costs (25c per day) and some time to render them (I queue them overnight and wake up to 200 4k images to go through and select the best).
I have a lot of experience in automating workflows and finally found what works for me, so I'm experimenting with this. It's working well so far and I will continue.
Like with everything in life - the harder it is to make, the more value it has. Everyone can create images with midjourney. They all look the same and discord seems harder to automate. I haven't tried, so that's an assumption. Anyway, I would recommend to everyone to start producing genAI images that others can't easily replicate with midjourney. The more technical knowledge is needed (custom LORAs, nodes, prompting techniques), less people will be able to create it. I'm not concerned with sharing this information because I know how hard it is to make content like that, and 0.1% of people have the technical know-how of to do it. So even if I wrote it down step-by-step, most people wouldn't do it. To cut it short - yes, great to have a niche, but it has to be difficult to recreate and there has to be ample demand for it.
-
Team Her Ugliness ;D ;D ;D
-
I have been doing this for nearly 20 years.
Woooooow !
Nearly 20 years ? And your photos still look like they are created with mobile phone for the social media story post : "look what I just eat for breakfast" ?
Well...
-
You are making me a better ai artist...:)
AI ARTIST ? ;D ;D ;D ;D ;D
If someone who uses Ai is artist....person with a remote control is movie director. :)
-
...
Yea, but what does it matter? Public, not public. I have still earned much more money with AI than you and only spent a fraction.
Will I need to be able to file a dmca takedown complaint? Why? Why should I care. I have zero personal attachment to these images, I spent zero effort on them (expect for the keywords, which is the one part that is not public on Midjourney) and I hardly spent any money on them. A fraction of a cent per image a best. I really do not care. Unlike with my real actual photos, where I spend countless of hours of work and also money on. That's where I care. That's where I do not want my work to be stolen (which AI has basically done!!!!). I did not create these AI images! An AI did! I don't even feel like they should belong to me.
Microstock is about earning money. And if, at the bottom line, no matter whether "you will get back your money" or "images are public" or whatever else argument you try to bring forth, someone else is earning much more money than you, it all does not matter - The method of the person who earns more money is better than yours.
The ratio of how much you have spent on AI and what little you have earned back is insane! And you constantly use your own experience to explain how AI will not make real photographers pointless, how it will not bring down the microstock business and how not everyone can do it. How it was hard and took effort and money. When all of this is simply not true. Your method is simply not efficient.
You do not understand how EASY and CHEAP and FAST everyone could now make money with AI images.
Because you have not figured out for yourself how it works yet.
And once everyone finally figures it out no one will be able to earn any decent money anymore, because the pie is not big enough. Expect the companys who bulit the AIs. The only ones who will in the long run profit from this.
exactly - why spend time whinging about declining sales & following futile pathways when you could be making money with AI gen. i have an enormous backlog of images from my travels but mostly low priority. instead i'm working improve my AI images rather than submitting images that are already over-represented
-
6 days after crossing the 14 million threshold, Adobe Stock's genAI collection has topped 14.5 million - 14,502,719 this morning
-
I have been doing this for nearly 20 years.
Woooooow !
Nearly 20 years ? And your photos still look like they are created with mobile phone for the social media story post : "look what I just eat for breakfast" ?
Well...
But how dare you talk to a person like that?
And who do you think you are?Do you think you are better than someone else?
But no one has taught you a minimum of respect?
-
I have been doing this for nearly 20 years.
Woooooow !
Nearly 20 years ? And your photos still look like they are created with mobile phone for the social media story post : "look what I just eat for breakfast" ?
Well...
But how dare you talk to a person like that?
And who do you think you are?Do you think you are better than someone else?
But no one has taught you a minimum of respect?
Thank you, these thoughts were also in my head - but I do not always want to play here the etiquette and decency watchdog ;)
Besides, mobile phone food photos from everyday life sell very well and are made quickly :D
-
Looking at today's new images in Adobe Stock's genAI collection is just depressing - it's littered with "oops" images - extra hand, objects malformed or missing bits or floating. Then there are multi-level wind turbines (which I predict will never be a thing, at least as pictured)
I'll call this the "We don't give a $h1t" collection
(https://t3.ftcdn.net/jpg/06/31/99/60/240_F_631996001_CcsdjCWQfQQUaCktPDKpQpWKGW4u5w8B.jpg) (https://stock.adobe.com/images/realistic-photo-of-elderly-man-cooking-in-domestic-kitchen-by-generative-ai/631996001)
(https://t4.ftcdn.net/jpg/06/35/23/55/240_F_635235532_RagG9UXxBfdIXvBLnRqyREyRO6y9Gtum.jpg) (https://stock.adobe.com/images/typewriter-with-fall-autumn-leaves-on-a-dark-background-top-view-ai-generated/635235532)
(https://t4.ftcdn.net/jpg/06/32/59/81/240_F_632598173_ZqIuNyCFMABXFCF7rjlngtYiAQdFSKsQ.jpg) (https://stock.adobe.com/images/overhead-shot-of-lemons-with-visible-water-drops-close-up/632598173)
(https://t4.ftcdn.net/jpg/06/32/61/95/240_F_632619549_dcGs84Fjp03AGvjrJFYyJ1bAVgFajFlp.jpg) (https://stock.adobe.com/images/white-bathroom-interior/632619549)
(https://t4.ftcdn.net/jpg/06/29/41/93/240_F_629419396_n9p6OCNCT40tObJ8uE7SdLWUuwGfFxJR.jpg) (https://stock.adobe.com/images/water-or-alcohol-in-a-glassful-with-a-lemon/629419396)
(https://t4.ftcdn.net/jpg/06/29/43/51/240_F_629435187_a4Qmq6OYAy6BF4YMoA9l1gyfL8cchni7.jpg) (https://stock.adobe.com/images/wind-generators-on-a-cloudy-ridge/629435187)
-
You are making me a better ai artist...:)
AI ARTIST ? ;D ;D ;D ;D ;D
If someone who uses Ai is artist....person with a remote control is movie director. :)
Loghorrea Artist, that's for sure, but for the rest? I feel incompetent to assess... ;D
-
I'll call this the "We don't give a $h1t" collection
Good finds! I've posted 3 of them in the AS Discord #qualitycontrol channel.
-
Most of you bow down to praise the AI and enjoy to use it. Generating tons of AI images leaves you too much free time? Why complain then? do something else, watch other things, spend your time elsewhere and don't use AI. And stop promoting this nightmarish dream.
[edit]: Do you seriously think you have leverage to influence Adobe? They trace their path, a path that will narrow greatly for humans to whom they must donate part of their profits...
-
I have been doing this for nearly 20 years.
Woooooow !
Nearly 20 years ? And your photos still look like they are created with mobile phone for the social media story post : "look what I just eat for breakfast" ?
Well...
But how dare you talk to a person like that?
And who do you think you are?Do you think you are better than someone else?
But no one has taught you a minimum of respect?
How do you dare to speak to Madoo in this way? Who are you and do you think you are better then Madoo?
Cobalt has a big mouth herself in her posts and she is surely not shy of insulting people. She is not an innocent cute little puppy to defend. She does almost only upload, indeed, pictures of her breakfast or other snacks she rans into. So the remark of Madoo is not at all ill placed.
The respect you think is missing with Madoo, is missing with you as well because you don't mind of having a big insulting mouth yourself to another person, do you?
First look into the mirror and then judge someone else :)
You have had a terrible upbringing yourself and please stop with your needless political correct, saviour like, replies. Doesn't make it any sense at all even if Ralf believes so (same upbringing, I guess)
-
I have been doing this for nearly 20 years.
Woooooow !
Nearly 20 years ? And your photos still look like they are created with mobile phone for the social media story post : "look what I just eat for breakfast" ?
Well...
But how dare you talk to a person like that?
And who do you think you are?Do you think you are better than someone else?
But no one has taught you a minimum of respect?
How do you dare to speak to Madoo in this way? Who are you and do you think you are better then Madoo?
Cobalt has a big mouth herself in her posts and she is surely not shy of insulting people. She is not an innocent cute little puppy to defend. She does almost only upload, indeed, pictures of her breakfast or other snacks she rans into. So the remark of Madoo is not at all ill placed.
The respect you think is missing with Madoo, is missing with you as well because you don't mind of having a big insulting mouth yourself to another person, do you?
First look into the mirror and then judge someone else :)
You have had a terrible upbringing yourself and please stop with your needless political correct, saviour like, replies. Doesn't make it any sense at all even if Ralf believes so (same upbringing, I guess)
SVH, it's good to see that you've regained your old form in appropriate human social interaction - that must feel really good for you.
You may have forgotten - it can happen - that you have been kicked out of the forum already before for bad behavior
and have bullied and insulted other members here.
Such personal attacks against other forum members are completely unnecessary. That is not really an enrichment for the discussions and mood in the forum.
-
....
SVH, it's good to see that you've regained your old form in appropriate human social interaction - that must feel really good for you.
You may have forgotten - it can happen - that you have been kicked out of the forum already before for bad behavior
and have bullied and insulted other members here.
Such personal attacks against other forum members are completely unnecessary. That is not really an enrichment for the discussions and mood in the forum.
yes, some of the personal attacks are getting close to being ban-worthy
thanks for trying to get this thread back on track
-
I have been doing this for nearly 20 years.
Woooooow !
Nearly 20 years ? And your photos still look like they are created with mobile phone for the social media story post : "look what I just eat for breakfast" ?
Well...
But how dare you talk to a person like that?
And who do you think you are?Do you think you are better than someone else?
But no one has taught you a minimum of respect?
How do you dare to speak to Madoo in this way? Who are you and do you think you are better then Madoo?
Cobalt has a big mouth herself in her posts and she is surely not shy of insulting people. She is not an innocent cute little puppy to defend. She does almost only upload, indeed, pictures of her breakfast or other snacks she rans into. So the remark of Madoo is not at all ill placed.
The respect you think is missing with Madoo, is missing with you as well because you don't mind of having a big insulting mouth yourself to another person, do you?
First look into the mirror and then judge someone else :)
You have had a terrible upbringing yourself and please stop with your needless political correct, saviour like, replies. Doesn't make it any sense at all even if Ralf believes so (same upbringing, I guess)
SVH, it's good to see that you've regained your old form in appropriate human social interaction - that must feel really good for you.
You may have forgotten - it can happen - that you have been kicked out of the forum already before for bad behavior
and have bullied and insulted other members here.
Such personal attacks against other forum members are completely unnecessary. That is not really an enrichment for the discussions and mood in the forum.
I only repeated an other mans words. What harm can that do? I would have expelled you a lot sooner. Mirror, mirror. You should have a look at your own comments. But probably blind to see your own faults.
-
....
SVH, it's good to see that you've regained your old form in appropriate human social interaction - that must feel really good for you.
You may have forgotten - it can happen - that you have been kicked out of the forum already before for bad behavior
and have bullied and insulted other members here.
Such personal attacks against other forum members are completely unnecessary. That is not really an enrichment for the discussions and mood in the forum.
yes, some of the personal attacks are getting close to being ban-worthy
thanks for trying to get this thread back on track
No, the thread is about AI generated photos. Please carry on, I was just reacting to political correct persons here that went off topic.
-
Like with everything in life - the harder it is to make, the more value it has. Everyone can create images with midjourney. They all look the same and discord seems harder to automate. I haven't tried, so that's an assumption. Anyway, I would recommend to everyone to start producing genAI images that others can't easily replicate with midjourney. The more technical knowledge is needed (custom LORAs, nodes, prompting techniques), less people will be able to create it. I'm not concerned with sharing this information because I know how hard it is to make content like that, and 0.1% of people have the technical know-how of to do it. So even if I wrote it down step-by-step, most people wouldn't do it. To cut it short - yes, great to have a niche, but it has to be difficult to recreate and there has to be ample demand for it.
Couldn‘t agree more.
When I started to do ai for stock, I wanted to experiment with genres I have never done and lots of different art styles, oil paintings, charcoal, sketches …. things you cannot do with a camera.
Obviously…this is not highly needed content. It is a lot easier to make money taking pictures of ugly breakfast with an old mobile phone. Gets very reliable sales too.
Robots and fantasy are also completely new things.
But I didn‘t research them, in the beginning it was all just learning. And I wanted to do something that is as far away from photography as possible.
I admire that you can do it all on your own, certainly cheaper and absolutely private.
I like a lot of what I see online done by midjourney, but I don’t like what people are uploading to stock agencies. Very few people do something really interesting.
However I am sure this will evolve and there will be some really amazing visuals coming from a new generation of artists.
Costs have come down a lot for me, because now I have a little library of prompt basics for some genres.
Still another three themes I want to test and then I will have my 1200 ai files and I will learn from the customers what they like to buy.
Will it ever be as easy to make money as ugly social media food? Or easy travel and nature shots?
I doubt it. But it brings diversity and these are the early stages.
ai is here to stay and we will all still do it in 30 years.
eta
and if it all does not work out…another day, another breakfast…
-
I have been doing this for nearly 20 years.
Woooooow !
Nearly 20 years ? And your photos still look like they are created with mobile phone for the social media story post : "look what I just eat for breakfast" ?
Well...
But how dare you talk to a person like that?
And who do you think you are?Do you think you are better than someone else?
But no one has taught you a minimum of respect?
Why ignore that there can be people more competent than others? oh yes, with AI generation, everybody's equal in term of skills, that is a big change... for others ;)
Are you writing this against Madoo or cobalt? I'm not sure that the person you denounce as being pretentious is the right one here. Which of the two believes and defines herself as an expert?
“Ignorance more frequently begets confidence than does knowledge...” Charles Darwin
-
Well I suppose I have been labelled as the expert in ugly social media breakfast food. But it is not my main income, that was just what sold on eyeem.
It's ok, people who cannot even make money with their breakfast...
I am sure they are superbly brilliant and everything they create is an instant top seller. No need to experiment and try things out. Especially not ai ;)
eta: breakfast and lunch gave me 3-6 files today, including one handheld video.
I will be rich when I retire on my daily food musings.
-
Looking at today's new images in Adobe Stock's genAI collection is just depressing - it's littered with "oops" images - extra hand, objects malformed or missing bits or floating. Then there are multi-level wind turbines (which I predict will never be a thing, at least as pictured)
I'll call this the "We don't give a $h1t" collection
Adobe obviously outsourced the bulk of their reviewing to India, Pakistan, Vietnam and the surrounding countries.
-
I don't know who or where reviewing is happening, but the results would be funny if it wasn't undermining stock images as a useful resource
(https://t4.ftcdn.net/jpg/06/32/29/65/240_F_632296556_5S9ATQiKqF0YJ9UXYzKY34KzRFR2vmjR.jpg) (https://stock.adobe.com/images/many-glasses-and-glasses-with-multicolored-appetizing-fruit-cocktails-and-juices-on-black-drink-background-for-advertising-and-bar-design/632296556)
-
I'll call this the "We don't give a $h1t" collection
Good finds! I've posted 3 of them in the AS Discord #qualitycontrol channel.
Thanks for posting. I note that they're all still there morning of 20 Aug, so possibly these aren't viewed as a problem any more? Or everyone at Adobe's hair is on fire because of AI content imitating the style of copyrighted works?
-
Many of the images on the Midjourney showcase are with texts ‘in the style of’ followed by artist name, or sometimes directly ‘by artist’, so it’s probably not a surprise. I suppose some of them uploading them to stock sites won’t include those names in their titles and tags.
-
Thanks for posting. I note that they're all still there morning of 20 Aug, so possibly these aren't viewed as a problem any more? Or everyone at Adobe's hair is on fire because of AI content imitating the style of copyrighted works?
Adobe doesn't remove the files themselves. Apparently the method they are using to clean up the Gen AI content is that firstly Adobe blocks the contributor, then the contributor has to write to support and get informed of the reason for their block, then they are given their account back so long as they agree to remove all problematic content within the time provided. (I have no idea how long they get.) If they do not do so in the time provided, they will get blocked again, or ......terminated. So when images that have been reported are still online there it's probably because that person's account has been left open for them to clean it up.
-
I don't know who or where reviewing is happening, but the results would be funny if it wasn't undermining stock images as a useful resource
Well I can tell you for sure that the reviewers are not US-based
How do I know? I'm checking the # of images in the database on an increased frequency, and there's a drop in the "images reviewed per hour" metric during the weekend (for obvious reasons), which disappeared overnight. When it was between midnight and 8am in the US, the images reviewed per hour picked up to normal weekday speed. So it's either Europe or the countries I mentioned. Due to the costs associated with this and the ease of offshoring, I bet on the latter.
-
Thanks for posting. I note that they're all still there morning of 20 Aug, so possibly these aren't viewed as a problem any more? Or everyone at Adobe's hair is on fire because of AI content imitating the style of copyrighted works?
Adobe doesn't remove the files themselves. Apparently the method they are using to clean up the Gen AI content is that firstly Adobe blocks the contributor, then the contributor has to write to support and get informed of the reason for their block, then they are given their account back so long as they agree to remove all problematic content within the time provided. (I have no idea how long they get.) If they do not do so in the time provided, they will get blocked again, or ......terminated. So when images that have been reported are still online there it's probably because that person's account has been left open for them to clean it up.
That is a pretty smart strategy to "educate" producers and make them clean up their own mess. And if they don't drop their bad habits, they have to leave.
Not all people are ill intentioned, many are just ill informed. Especially if their first language is not English and some dubious youtuber or social media guru is "teaching " them the quick way to instant millions via Adobe.
-
It's a little scary though. What if your not very skilled or knowledgeable regarding generating AI content and you didn't do anything intentionally problematic. Your account gets shut down? Shouldn't it be the inspectors job not to accept such content? What if you are a new photographer and your pics are not in focus or have some other technical issue. Your photos won't get accepted...that's fine....but the analogous situation would be if AS accepted out of focus images but then later shut down your account.....because they didn't reject your out of focus photos in the first place?
-
It's a little scary though. What if your not very skilled or knowledgeable regarding generating AI content and you didn't do anything intentionally problematic. What if you are a new photographer and your pics are not in focus or have some other technical issue. Your photos won't get accepted...that's fine....but the analogous situation would be if AS accepted out of focus images but then later shut down your account.....because they didn't reject your out of focus photos in the first place?
Comparing not being able to produce images of high quality and therefore not getting your images accepted and breaking rules are completely different things.
There is no rule on Adobe that "forbids" you to submit images that are not in focus. There is now a rule that forbids you to submit AI images that are labaled as having used someone else's style. If you want to earn money with AI images it is YOR responsibility to read up about rules and restrictions both of the AI creator you are using and the agency you are submitting your images to. That's really the minimum of effort you have to take. "not being very knowledgeable" is no excuse, because the knowledge is out there, you have to read.
-
It's a little scary though. What if your not very skilled or knowledgeable regarding generating AI content and you didn't do anything intentionally problematic. What if you are a new photographer and your pics are not in focus or have some other technical issue. Your photos won't get accepted...that's fine....but the analogous situation would be if AS accepted out of focus images but then later shut down your account.....because they didn't reject your out of focus photos in the first place?
Comparing not being able to produce images of high quality and therefore not getting your images accepted and breaking rules are completely different things.
There is no rule on Adobe that "forbids" you to submit images that are not in focus. There is now a rule that forbids you to submit AI images that are labaled as having used someone else's style. If you want to earn money with AI images it is YOR responsibility to read up about rules and restrictions both of the AI creator you are using and the agency you are submitting your images to. That's really the minimum of effort you have to take. "not being very knowledgeable" is no excuse, because the knowledge is out there, you have to read.
Valid points, but why would such an AI image pass inspection in the first place?...plus the rules are changing in real time.
-
This is a historic change in art, maybe even the biggest in the entire history of human art.
There are no fixed long proven rules. Everything is evolving in real time.
Keep in mind that Adobe is the only agency that is giving us the opportunity to make money with ai images (ok, dreamstime takes them too, but few sales).
And they also give customers a beautiful new collection to browse. Yes, there are errors, but it is a process for all sides involved.
Of course the criminals (or misinformed) are taking the opportunity to evade the rules, spurned on by wild social media influencers promising easy money.
Personally I think for the situation Adobe is working at lightspeed under the circumstances.
-
And this number of 9 million has crossed 18 million by today viz. in approx 4 months
-
And this number of 9 million has crossed 18 million by today viz. in approx 4 months
So is that about 10% of the collection?
-
Now that Adobe uses those images to train Firefly, I can understand the rush to create a huge Ai collection.
-
I see that AI has allowed many low-quality contributors to finally submit high-quality content.
Contributors with 40,000 bad images who struggled to make $100 a month now think they're making a lot of money.
But for me, it's a mistake. Sure, maybe they will make a little more money. Because AI has allowed us to raise the quality of jobs, but now they are all the same.
I struggle to distinguish the work of one contributor from another.
They are all identical, all with the same style.
And so in the end everyone will earn quite similar amounts. There will be a race to see who can upload the most.
And I think this will eventually tire customers who want to find something different from this mountain of boring images copied from artists and photographers of the past.
I think differentiating yourself from others will be the most important thing in the future.
I don't know if we must do it with generic stock agencies or niche agencies, but earning a full salary with AI will be very difficult. There are too many contributors and too many images for the same pie to share.
-
I don't know if we must do it with generic stock agencies or niche agencies, but earning a full salary with AI will be very difficult. There are too many contributors and too many images for the same pie to share.
There will be no earning any ‘salary’ from AI images for contributors, clients will type what they want and pay to the big companies. At least that’s how I see it.
-
I don't know if we must do it with generic stock agencies or niche agencies, but earning a full salary with AI will be very difficult. There are too many contributors and too many images for the same pie to share.
There will be no earning any ‘salary’ from AI images for contributors, clients will type what they want and pay to the big companies. At least that’s how I see it.
If you try to generate what you exactly want you'll see how difficult that is, so for the clients is much more easy and less time consuming to use a good AI search engine and find the best of AI generated content from image data base. To me the future is in the best search engine.
-
I don't know if we must do it with generic stock agencies or niche agencies, but earning a full salary with AI will be very difficult. There are too many contributors and too many images for the same pie to share.
There will be no earning any ‘salary’ from AI images for contributors, clients will type what they want and pay to the big companies. At least that’s how I see it.
If you try to generate what you exactly want you'll see how difficult that is, so for the clients is much more easy and less time consuming to use a good AI search engine and find the best of AI generated content from image data base. To me the future is in the best search engine.
I agree with you Mir. I didn't write it in my post, but what you say is true. Many customers will create their own custom images.
When I talked about salary, I was referring to those who do photography or illustration in a traditional way, without AI. The key is to differentiate yourself and communicate well with a blog, a YouTube channel, or a podcast. Often, those who buy images fall in love with the author's story or the process of creating the images.
I don't see a great future in uploading images to the big stock agencies. As for photography and illustration authors, only the best will survive. Maybe I'm wrong, but I believe we need to invest time in improving the quality of our work, not in producing millions of useless AI images. Agencies will profit and build their own AI engine. But we contributors don't, the slices to be shared are too small.
-
If you try to generate what you exactly want you'll see how difficult that is, so for the clients is much more easy and less time consuming to use a good AI search engine and find the best of AI generated content from image data base. To me the future is in the best search engine.
1. No, it's not.
2. It's most certainly not more difficult than getting "exactly what you want" when searching for an image in a Ai image / microstock agency.
-
I agree with you Mir. I didn't write it in my post, but what you say is true. Many customers will create their own custom images.
When I talked about salary, I was referring to those who do photography or illustration in a traditional way, without AI. The key is to differentiate yourself and communicate well with a blog, a YouTube channel, or a podcast. Often, those who buy images fall in love with the author's story or the process of creating the images.
I don't see a great future in uploading images to the big stock agencies. As for photography and illustration authors, only the best will survive. Maybe I'm wrong, but I believe we need to invest time in improving the quality of our work, not in producing millions of useless AI images. Agencies will profit and build their own AI engine. But we contributors don't, the slices to be shared are too small.
My thinking is in line with what you wrote above, I don't think that non AI images will stop selling, but only a few will be able to continue making a living with it. My previous comment wasn't sarcastic or anything, sorry if it sounded like that.
If you try to generate what you exactly want you'll see how difficult that is, so for the clients is much more easy and less time consuming to use a good AI search engine and find the best of AI generated content from image data base. To me the future is in the best search engine.
What Her Ugliness said.
-
Made 114.83 USD so far
But I have paid over 2000 dollars in credits for ai use
If you are spending so much more than what you are earning that makes you a bad example how not to use AI for microstock. :o
I've only spent $71.40 on AI and earned back a multiple of it. If I hadn't, I would stop spending money on AI.
My son spent the same exactly (71.4$) and earned 284$ in the last 5 months.
how have your son's revenues and downloads changed since your last review?
-
And this number of 9 million has crossed 18 million by today viz. in approx 4 months
So is that about 10% of the collection?
More than 5%
-
There are only 2.4 million AI generated photos on Adobe Stock. That’s roughly only 1% of total stock photos.
-
There are 18,401,010+ items in the genAI collection, of which 2,428,094+ are tagged as photos, but many more than 2 million are photo-realistic. For quite a while contributors were required to submit genAI content as an illustration.
There are 365+ million items on Adobe Stock, 209+ million of which are photos, the biggest subset.
But as long as idiot "photos" like this (from today) are accepted, you have to wonder if we need another category - along with better keywords to cover the situation if you really want to depict a 3-legged woman:
"Beautiful young woman sitting in folding chair"
(https://t3.ftcdn.net/jpg/06/55/96/68/240_F_655966802_OxqpAJybdxzfAkaja9pz9ISsualRwZo1.jpg) (https://stock.adobe.com/images/beautiful-young-woman-sitting-in-folding-chair/655966802)
Keywords are rubbish such as forest, notebook, cyberspace, technology, along with useful ones such as woman, grass, outside, folding chair, etc.
This is neither a photo, nor an illustration, even if the 3 legs were on purpose.
-
Is this the same person with three different accounts or are they just copying the same title.
Charity illustration concept with abstract, diverse persons, hands and hearts. Community compassion, love, and support towards those in need.
(https://t4.ftcdn.net/jpg/06/10/95/69/240_F_610956975_79mPdRcXcR9ikwBAIRiktwpE7rjbKgrb.jpg) (https://stock.adobe.com/se/images/charity-illustration-concept-with-abstract-diverse-persons-hands-and-hearts-community-compassion-love-and-support-towards-those-in-need/610956975?as_campaign=ftmigration2&as_channel=dpcft&as_campclass=brand&as_source=ft_web&as_camptype=acquisition&as_audience=users&as_content=closure_asset-detail-page)
(https://t3.ftcdn.net/jpg/05/91/42/68/240_F_591426890_RZBHC4vqNaOuhNWZ45f6KEAX91FQRuoY.jpg) (https://stock.adobe.com/images/charity-illustration-concept-with-abstract-diverse-persons-hands-and-hearts-community-compassion-love-and-support-towards-those-in-need-generative-ai/591426890?as_campaign=ftmigration2&as_channel=dpcft&as_campclass=brand&as_source=ft_web&as_camptype=acquisition&as_audience=users&as_content=closure_asset-detail-page)
(https://t3.ftcdn.net/jpg/06/07/38/02/240_F_607380255_Uthxq0iNvOwacwOZvIbiqosxj8P32Wtr.jpg) (https://stock.adobe.com/fr/images/charity-illustration-concept-with-abstract-diverse-persons-hands-and-hearts-community-compassion-love-and-support-towards-those-in-need/607380255?as_campaign=ftmigration2&as_channel=dpcft&as_campclass=brand&as_source=ft_web&as_camptype=acquisition&as_audience=users&as_content=closure_asset-detail-page)
-
'Funny' .. was three legs part of the keywords ..? ;D
I don't think the agencies even check keywords anymore except for banned terms. Apparently it's also quite easy to get a lot of photos accepted at Adobe via AI - see the three here. And strange hands and a lot of wristwatches ...
-
After 12 years selling and 9 years full-time producing stock art I'm totally out. Deleted my last image this week.
Last week I got my mower serviced for $580. 2hrs labour and parts. It would take hundreds of sales of stock images to pay for that service and the people buying the art use it for years. They should be paying $40 to use it for 10 years and we should get at least 40% of the sale.
Opaque agencies sell art for a dollar and give the artist a few cents. Taking portfolio rights and training their AI systems without consultation. It's the most ridiculous business now. We have people just using AI to make new images as an intermediary step until AI does it all.
I have no more art online unless of course it turns up in some AI Frankenstein creation. I probably won't renew my Adobe subscription after more than 30 years of using their software. Bye.
-
9 Million+ AI generated photos - Stock Photography coming to end
Unfortunately, yes!
Just like film photography, vinyl record players, tape recorders, paper books and newspapers, corded telephones, optical discs and much more.
And stock photography is killed by production studios, united teams of stockers, and even single stockers because of their own greed and stupidity.
It doesn’t matter that artificial intelligence still takes clumsy pictures. It won't be long before he learns. And we will be "the last of the Mohicans." And yes, indeed, we will have to carry lawn mowers. But, however, this is only if by that time the lawns are not mowed by biorobots.
-
9 Million+ AI generated photos - Stock Photography coming to end
Unfortunately, yes!
Just like film photography, vinyl record players, tape recorders, paper books and newspapers, corded telephones, optical discs and much more.
And stock photography is killed by production studios, united teams of stockers, and even single stockers because of their own greed and stupidity.
It doesn’t matter that artificial intelligence still takes clumsy pictures. It won't be long before he learns. And we will be "the last of the Mohicans." And yes, indeed, we will have to carry lawn mowers. But, however, this is only if by that time the lawns are not mowed by biorobots.
Already here: https://buyersguide.org/lawn-mowers/t/robot?msclkid=3cfaf996b37c17edb8e79cbbd0501e2e&m=e&d=c&c=74011169756934&oid=kwd-74011059969351:loc-4133&qs=robot%20lawn%20mower&lp=105822&li=&nw=o&nts=1&tdid=8844783
Of all the things you listed, the one that seems to be surviving is paper books. For some reason, people still enjoy a good book, rather than a computer screen. All the rest have been made obsolete by newer technology that does a better job. I'm not convinced that AI/Machine Learning does a better job. I'm not hoping that it does for a while.
A computer can do some things, that humans have taught or programmed, but it can't reason or know function, or why. That's why 3 arms, 7 toes, or mechanical things, just get mashed into flawed and often impossible combinations.
-
I would like to add to Pete's contribution that for about 10 years, at least in Germany, sales of vinyl records and players have been growing steadily.
Jazz goes for me only on vinyl and that seems to be the trend ;)
Not everything is lost. :D
-
...
A computer can do some things, that humans have taught or programmed, but it can't reason or know function, or why. That's why 3 arms, 7 toes, or mechanical things, just get mashed into flawed and often impossible combinations.
a bit too optimistic - we already have ai translators which do not understand the languages they render - world class chess & go & Jeopardy apps that don't even know the rules of their games and many more. for now these work for specific problems but that will expand to include tasks now considered to require human consciousness. remember that once humans were defied by being the only species who used language
earle (1999) summarized his Chinese Room Argument concisely:
Imagine a native English speaker who knows no Chinese locked in a room full of boxes of Chinese symbols (a data base) together with a book of instructions for manipulating the symbols (the program). Imagine that people outside the room send in other Chinese symbols which, unknown to the person in the room, are questions in Chinese (the input). And imagine that by following the instructions in the program the man in the room is able to pass out Chinese symbols which are correct answers to the questions (the output). The program enables the person in the room to pass the Turing Test for understanding Chinese but he does not understand a word of Chinese.
https://plato.stanford.edu/entries/chinese-room/
-
I would like to add to Pete's contribution that for about 10 years, at least in Germany, sales of vinyl records and players have been growing steadily.
Jazz goes for me only on vinyl and that seems to be the trend ;)
Not everything is lost. :D
and after 150 years of photography we still have artists who smear colored liquids on various surfaces, though few in the photorealistric style of David, Coubier, Daumier et al - instead it led to the expanding conciousness of impressionism, expressionism etc
-
Of all the things you listed, the one that seems to be surviving is paper books. For some reason, people still enjoy a good book, rather than a computer screen. All the rest have been made obsolete by newer technology that does a better job. I'm not convinced that AI/Machine Learning does a better job. I'm not hoping that it does for a while.
A computer can do some things, that humans have taught or programmed, but it can't reason or know function, or why. That's why 3 arms, 7 toes, or mechanical things, just get mashed into flawed and often impossible combinations.
So the newspapers are still preserved. The point is not that they have been preserved - but in their quantity. When we have elections, the number of newspapers increases sharply. This is how candidates fight for the votes of pensioners :) I specifically go around the city and collect newspapers for a collection. The elections are over - newspapers are reduced to a minimum.
Don't criticize artificial intelligence - it just started elementary school. When he graduates, then we’ll see what he can do.
I would like to add to Pete's contribution that for about 10 years, at least in Germany, sales of vinyl records and players have been growing steadily.
Wow, wow! And who produces them?
My old supply of radio tubes was sold out, and the buyers almost tore my arms off. And they asked again - but I don’t have any more lamps :) Is some factory producing radio tubes now?
-
I would like to add to Pete's contribution that for about 10 years, at least in Germany, sales of vinyl records and players have been growing steadily.
Wow, wow! And who produces them?
Big record companies. I am also from Germany and I can confirm. :) I do not know whether it's all music genre, but I listen to metal and here every metal band that brings out a new album will also release multiple versions of it on vinyl. And metal fans swear that they just sound so much better than CDs - and mp3s of course.
And, as a book lover with over 2000 books I can also confirm that paper books are still going strong. And I think one of the selling points of books is that people like to give them to other people as gifts. And that just doesn't work so well with e-book files. ;)
But I do not think that this concept can be applied well to the AI vs. real photos discussion, because the problem here is that in many cases you cannot tell them apart. I mean, I think I still can in many cases, but I have seen people fall for AI images online, thinking they were real, that looked so obviously AI generated to me. So many people can't tell the difference, probably especially people who have never taken real photos other than with their phones and have never generated AI images. But I think as AI images will advance further, it will become more and more difficult to tell them apart from real photos, even to the trained eye. And then the comparison to vinyls or books just doesn't work anymore, because you can obviously easily tell these apart from their modern digital counterparts. But how do you chose between an AI image and a real photo as a customer when you just do not know which one is which?
-
I am also from Germany and I can confirm.
So, can you confirm, I’m very interested.
I didn’t deny that there are gourmets who adore vintage things. A friend of mine drives a 1939 BMW. And what?
And metal fans swear that they just sound so much better than CDs
Well, yes, well, yes... And the gramophone sounds even better. ;D
But I do not think that this concept can be applied well to the AI vs. real photos discussion, because the problem here is that in many cases you cannot tell them apart. I mean, I think I still can in many cases, but I have seen people fall for AI images online, thinking they were real, that looked so obviously AI generated to me. So many people can't tell the difference, probably especially people who have never taken real photos other than with their phones and have never generated AI images. But I think as AI images will advance further, it will become more and more difficult to tell them apart from real photos, even to the trained eye. And then the comparison to vinyls or books just doesn't work anymore, because you can obviously easily tell these apart from their modern digital counterparts. But how do you chose between an AI image and a real photo as a customer when you just do not know which one is which?
Absolutely right! And the camera will become a cool trinket from the past. But there will be gourmets who will continue to take photographs. And it's sad and wonderful at the same time.
-
I am also from Germany and I can confirm.
So, can you confirm, I’m very interested.
I didn’t deny that there are gourmets who adore vintage things. A friend of mine drives a 1939 BMW. And what?
But it's a bit different with vinyls. A 1939BMW is a rarity and your friend driving one is an exception to what is "normal", but vinyls are not such a rarity.
https://www.theguardian.com/music/2023/jul/12/vinyl-sales-us-report (https://www.theguardian.com/music/2023/jul/12/vinyl-sales-us-report)
Absolutely right! And the camera will become a cool trinket from the past. But there will be gourmets who will continue to take photographs. And it's sad and wonderful at the same time.
Yes, people will continue to take photos. Especially since AI cannot replace some things - not photos of your son's birthday party, not your wedding, not your pet. And some people will still do it as a hobby. But will people still be able to make a living from photos in the microstock market? I have my doubts.
-
Paper books vs. digital books, and vinyl vs. digital formats is not the same as drawings/ photography by human beings vs. drawings/ photography by AI. In the first case the medium is changed but the author is the same, you can continue writing books and music and selling them in digital format, with art/ photography your input is replaced.
-
Yes, people will continue to take photos.
Undoubtedly!
But only those who, instead of one of their eyes, are implanted with an artificial one with a built-in high-resolution camera with the ability to connect to the Internet via Wi-Fi.
Paper books vs. digital books, and vinyl vs. digital formats is not the same as drawings/ photography by human beings vs. drawings/ photography by AI. In the first case the medium is changed but the author is the same, you can continue writing books and music and selling them in digital format, with art/ photography your input is replaced.
With the help of artificial intelligence, not only books are written, but doctoral dissertations are written.
Artificial intelligence will show everyone where crayfish spend the winter, don’t flatter yourself.
-
With the help of artificial intelligence, not only books are written, but doctoral dissertations are written.
Artificial intelligence will show everyone where crayfish spend the winter, don’t flatter yourself.
I only said that comparing paper and digital books is not the same as comparing art created by people and art created by AI.
Replace art with books, it is still valid.
-
Edited Tuesday to add that I haven't been looking at Freepik and didn't realize they've been accepting genAI content for many months. They now have 26,580,000 genAI assets, so they're bigger than Adobe Stock's genAI collection.
One interesting feature they have is a box under each genAI item for "Base model", and although many contributors just say "Not specified", I also saw Midjourney 5.2, Stable Diffusion 1.5, Stable Diffusion 2.0, Dall-e 1, Stable Diffusion 2.1, Leonardo. Midjourney seems to be the most used of the ones named.
Lots of logos in their collection too.
==========================
I have been tracking the size of the genAI collections each Monday.
Adobe Stock's is the largest by farsecond largest after Freepik; Dreamstime is the only other agency with any sizable collection. Shutterstock's nearly 1.4 million genAI images are mostly a cautionary tale of what can go horribly wrong (https://www.shutterstock.com/search?image_type=generated&sort=newest)- I don't know why they'd exhibit how bad their on-site AI tools are for all to see.
Adobe Stock's genAI collection is now 20,649,310 - up over 1 million from 16 Oct (19,590,366) which was up nearly a million from 9 Oct (18,641,502). Before that it was more like half a million per week.
Separately from the issue of usability of these images is where all the extra customers to buy these will come from?
At some point I'd have thought those new contributors drawn in by the AI gold rush would realize there wasn't much return and just drift away. Not this week though :)
Looking at the content overall, though there are some images only AI could do, the vast majority are just replicating well-covered subjects already in the human-created collection (and with tons of "similars" (https://stock.adobe.com/search?creator_id=287508&?&order=creation&k=lemon+leaves+background) which further dilutes sales)
Adobe Stock's collection is about half the size of Shutterstock's - but SS bulking up their numbers seemed to be to impress investors rather than anything useful for buyers. Adobe Stock reports 369,012,172 results in all. SS has "445,783,263 stock photos, 3D objects, vectors, and illustrations" but the footer of each page says "We have more than 734,000,000 images across Shutterstock's properties as of June 30, 2023"
-
I don't know why they'd exhibit how bad their on-site AI tools are for all to see.
:D
there are many questions still unanswered, which we will understand as we go.
For example,I have an all Adobe app plan,so I will have 1000 credits to generate.
with these 1000 credits how many images can I generate?How many of these generated images will be acceptable?How many of these images will be accepted for sale by Adobe Stock?I hope all of them,because I pay a subscription,I believe and hope this is the case but clearly I don't know yet!
Another thing to consider is that Adobe Stock subscribers will have 500 credits to generate,this is not good news for me,it would have been better if the credits were only available to Adobe software subscribers,or less credits for Adobe Stock customers,500 for a month are more than enough for a customer I think.
in any case we'll see,in a year or less,we'll all have clearer answers,I believe and hope that Adobe has considered all this,we'll see! :)
-
...
A computer can do some things, that humans have taught or programmed, but it can't reason or know function, or why. That's why 3 arms, 7 toes, or mechanical things, just get mashed into flawed and often impossible combinations.
a bit too optimistic - we already have ai translators which do not understand the languages they render - world class chess & go & Jeopardy apps that don't even know the rules of their games and many more. for now these work for specific problems but that will expand to include tasks now considered to require human consciousness. remember that once humans were defied by being the only species who used language
earle (1999) summarized his Chinese Room Argument concisely:
Imagine a native English speaker who knows no Chinese locked in a room full of boxes of Chinese symbols (a data base) together with a book of instructions for manipulating the symbols (the program). Imagine that people outside the room send in other Chinese symbols which, unknown to the person in the room, are questions in Chinese (the input). And imagine that by following the instructions in the program the man in the room is able to pass out Chinese symbols which are correct answers to the questions (the output). The program enables the person in the room to pass the Turing Test for understanding Chinese but he does not understand a word of Chinese.
https://plato.stanford.edu/entries/chinese-room/
Ah yes but I confess to being an optimist. One of my flaws, unless living in gloom and being a pessimist is just as ineffective?
I made a matchbox player, (AI or computer?) that could compete and not lose, at the game, in about 1960. Much like your example, the boxes didn't know the rules or the game and they didn't reason or think, they were just sliding drawers, where the decisions were learned and played.
Maybe I didn't state what I meant properly, or I'm still off somewhere else? :) AI doesn't reason, it doesn't know function, and that's why there are the flagrant errors that we see as obvious. I'm not claiming that AI can't be trained or can't learn, just that there are some distinct issues without being able to reason how gravity works, or spokes on a wheel, or that people only have two arms, and most only five fingers or five toes.
This one I saved because it was nearly enough to give someone a nightmare and at the same time, rather humorous. AI just really gone bad.
Just how does that seat work? Wouldn't it be a little uncomfortable? And what about the seat post? Oh My!
(https://i.postimg.cc/d0ZKVbNQ/bicycle-butt.jpg)
-
...
I made a matchbox player, (AI or computer?) that could compete and not lose, at the game, in about 1960. Much like your example, the boxes didn't know the rules or the game and they didn't reason or think, they were just sliding drawers, where the decisions were learned and played.
...
was that the one from Martin Gardner's SciAM column? I still have mine, saving it for display in my memorial library.
-
For those who haven't been following the collection sizes, Adobe Stock's genAI collection is almost at 23 million; Freepik's is over 30 million; Dreamstime's is a hair over 7 million.
Shutterstock's is such dreck it doesn't matter, but it's approaching 1.5 million
The "oops" images continue to be everywhere...
(https://img.freepik.com/premium-photo/photo-two-asian-student-kid-girl-boy-schoolchildren-brother-sister-smile-happy_584200-5293.jpg) (https://www.freepik.com/premium-ai-image/photo-two-asian-student-kid-girl-boy-schoolchildren-brother-sister-smile-happy_84531340.htm#fromView=recent&page=3&position=6&track=®ularType=ai)
(https://t3.ftcdn.net/jpg/06/57/07/88/240_F_657078853_2sONF4dTMuXAmbCXUDVRvfYLTdSSGG2h.jpg) (https://stock.adobe.com/images/hyper-realistic-bananas-in-cozy-market-setting/657078853) (https://t3.ftcdn.net/jpg/06/53/99/48/240_F_653994861_O2UHk4Iy30hDGSljGUUE4AMUIMCZAuFL.jpg) (https://stock.adobe.com/images/roasted-chicken-with-salad-and-other-food-items/653994861)
-
Another thing to consider is that Adobe Stock subscribers will have 500 credits to generate,this is not good news for me,it would have been better if the credits were only available to Adobe software subscribers,or less credits for Adobe Stock customers,500 for a month are more than enough for a customer I think.
in any case we'll see,in a year or less,we'll all have clearer answers,I believe and hope that Adobe has considered all this,we'll see! :)
But Adobe users also need these credits for generative fill or other generative uses within Photoshop.
https://helpx.adobe.com/firefly/using/generative-credits.html
So I wouldn't worry too much about losing sales to image generation. As we all know that is still very time consuming.
-
that AI photo of bananas definitely predates my 3310 banana photos sent to Adobe! :D
-
Another thing to consider is that Adobe Stock subscribers will have 500 credits to generate,this is not good news for me,it would have been better if the credits were only available to Adobe software subscribers,or less credits for Adobe Stock customers,500 for a month are more than enough for a customer I think.
in any case we'll see,in a year or less,we'll all have clearer answers,I believe and hope that Adobe has considered all this,we'll see! :)
But Adobe users also need these credits for generative fill or other generative uses within Photoshop.
https://helpx.adobe.com/firefly/using/generative-credits.html
So I wouldn't worry too much about losing sales to image generation. As we all know that is still very time consuming.
but would it have been better to have fewer credits for Adobe Stock subscribers?probably yes,now I don't think it makes much difference,but in the future could be this difference.
-
I have heard that freepik is now removing similar AI content as it is affecting the search.
-
Regarding AI images. I guess they may represent say 10% of the entire collection at Adobe numerically? However, I would venture to guess that "subjectwise" they could potentially be more of a threat than that 10% number suggests since in the AI world, having complex images with "models" and "sets" (e.g. business office, medical office) is a lot easier vs. "real world" photos where the photog would have to get actual models and actual "sets" or locations like offices etc.. Therefore I would expect the "10%" number to be misleading (no "snapshot" images with very low sales potential). In other words, there should be more of a sales impact than 10% might suggest. Having said all that, I haven't seen the impact sales-wise that this scenario would suggest.
I am made an awful lot of assumptions in my statements above. Does anyone care to comment?
-
Regarding AI images. I guess they may represent say 10% of the entire collection at Adobe numerically? However, I would venture to guess that "subjectwise" they could potentially be more of a threat than that 10% number suggests since in the AI world, having complex images with "models" and "sets" (e.g. business office, medical office) is a lot easier vs. "real world" photos where the photog would have to get actual models and actual "sets" or locations like offices etc.. Therefore I would expect the "10%" number to be misleading (no "snapshot" images with very low sales potential). In other words, there should be more of a sales impact than 10% might suggest. Having said all that, I haven't seen the impact sales-wise that this scenario would suggest.
I am made an awful lot of assumptions in my statements above. Does anyone care to comment?
I don't have any details about sales beyond sorting the genAI collection by downloads which would suggest that most of the the types of images it could uniquely do are not what contributors are uploading.
I looked at today's new genAI approvals and there are a bunch of rustic wood textures. The contributor has about 100 of those. What's the effing point? Same goes for views of fake-y nature scenes, food photos, abstract backgrounds, birds, puppies, kittens, etc. Even if it is a ton less work to set up food shots via AI, the collections already have tons of hot dog and hamburger pictures, so why add more?
One of the reasons I think the genAI content has stuck with the things already so well supplied is that many of the detailed settings that would work really well as stock and aren't already "full" are very hard for current genAI to get right - flawed details in factories, wind turbines, plane interiors, etc make these functionally useless to buyers.
-
I have no doubt that at the rate artificial intelligence is advancing, stock photography has a maximum of two years left to live. Right now the smart thing to do is to sell all the lenses, equipment and cameras on ebay before it is too late and they become worthless junk.
-
I have no doubt that at the rate artificial intelligence is advancing, stock photography has a maximum of two years left to live. Right now the smart thing to do is to sell all the lenses, equipment and cameras on ebay before it is too late and they become worthless junk.
don't make this mistake!
instead look for good real content that can override AI if you want,or start uploading AI content if you aren't already doing so.
then it's your life and it's your choice.
I see that you need a more "fresh" point of view because I know how difficult it can be to have dedicated so much and then see that today with a prompt everything can be done!It can be depressing and difficult to accept.
Don't give up!Instead change and adapt!
-
I have no doubt that at the rate artificial intelligence is advancing, stock photography has a maximum of two years left to live. Right now the smart thing to do is to sell all the lenses, equipment and cameras on ebay before it is too late and they become worthless junk.
I don't think it will be quite that bad.
Cameras won't die out. Hardly anyone will have their travel or wedding pictures recreated with AI, for example ;).
But it will of course hit some photographers hard. For example, I wonder whether real food photography still makes sense with all the effort involved.
As my portfolio mainly consists of real locations and editorials, I don't see myself at risk here for the time being and will keep my camera.
-
I have no doubt that at the rate artificial intelligence is advancing, stock photography has a maximum of two years left to live. Right now the smart thing to do is to sell all the lenses, equipment and cameras on ebay before it is too late and they become worthless junk.
This is the same thought of 99% of photographers years ago, at the time when digital photography grew up.
It's simply a description of how to NOT adapt yourself to change YOUR world, while the world around is changing without your consent
-
I have no doubt that at the rate artificial intelligence is advancing, stock photography has a maximum of two years left to live. Right now the smart thing to do is to sell all the lenses, equipment and cameras on ebay before it is too late and they become worthless junk.
I haven't sent any AI images, and don't think I ever will.
I get help from AI for my post-production, but I'm not interested in generative AI.
It's boring, and the images are all the same, all with the same style.
Producing 100 images with AI takes just two or three hours, so photography and illustration categories will soon be saturated with millions of images uploaded in no time.
I think eventually, people will get bored, and the majority of AI users will create their own images.
I hope in the future, there will always be a niche of customers who prefer real images captured with authentic landscapes, models, and objects. And I continue to work for that niche.
We'll see in a few years.
-
I have no doubt that at the rate artificial intelligence is advancing, stock photography has a maximum of two years left to live. Right now the smart thing to do is to sell all the lenses, equipment and cameras on ebay before it is too late and they become worthless junk.
I haven't sent any AI images, and don't think I ever will.
I get help from AI for my post-production, but I'm not interested in generative AI.
It's boring, and the images are all the same, all with the same style.
I think eventually, people will get bored, and the majority of AI users will create their own images.
I hope in the future, there will always be a niche of customers who prefer real images captured with authentic landscapes, models, and objects. And I continue to work for that niche.
We'll see in a few years.
This!
Agree 100%
-
Producing 100 images with AI takes just two or three hours?
do you mean to generate unfinished images,just prompts?yes,you can do even more in 2-3 hours,but the work is just started.
I just sent 10 AI images to Adobe and it took me 2 days!
I generated them with AI yes,but then I completely renovated,improved,added particular effects...it takes forever!
photographers and videographers will always be necessary,AI changes the game but cannot replace real content.
let's put ourselves in the customers' shoes:
I need to buy an image,a composition of a Carbonara pasta dish where i can see all ingredients needed for prepare the dish.
What do you think I should choose?
I believe that at least 80% of customers prefer to purchase a real image of a real Carbonara dish,you no?do you want a fake Carbonara dish?ok..you are in that 20%
all AI images are labeled,so there will be no possibility of confusion.
generative fill contents instead in my opinion are more like real content,even if unfortunately they must be labeled as AI,because it is probably not possible to quantify the level of AI used,even if the main subject remains real,it is labeled as AI.
i advise to all contributors to don't give up,then everyone has their own point of view,and obviously makes their own choices.
-
For those who haven't been following the collection sizes, Adobe Stock's genAI collection is almost at 23 million; Freepik's is over 30 million; Dreamstime's is a hair over 7 million.
Shutterstock's is such dreck it doesn't matter, but it's approaching 1.5 million
The "oops" images continue to be everywhere...
(https://img.freepik.com/premium-photo/photo-two-asian-student-kid-girl-boy-schoolchildren-brother-sister-smile-happy_584200-5293.jpg) (https://www.freepik.com/premium-ai-image/photo-two-asian-student-kid-girl-boy-schoolchildren-brother-sister-smile-happy_84531340.htm#fromView=recent&page=3&position=6&track=®ularType=ai)
(https://t3.ftcdn.net/jpg/06/57/07/88/240_F_657078853_2sONF4dTMuXAmbCXUDVRvfYLTdSSGG2h.jpg) (https://stock.adobe.com/images/hyper-realistic-bananas-in-cozy-market-setting/657078853) (https://t3.ftcdn.net/jpg/06/53/99/48/240_F_653994861_O2UHk4Iy30hDGSljGUUE4AMUIMCZAuFL.jpg) (https://stock.adobe.com/images/roasted-chicken-with-salad-and-other-food-items/653994861)
Poor little boy has three arms, one with a shriveled girls hand and the other with bad arthritis. Nice Godzilla bananas. The turkeys are bred that way, naturally. And these have passed inspection?
-
Producing 100 images with AI takes just two or three hours?
do you mean to generate unfinished images,just prompts?yes,you can do even more in 2-3 hours,but the work is just started.
I just sent 10 AI images to Adobe and it took me 2 days!
I'm no expert, but I see that many succeed.
They may not be the best images in the world but it is possible to take 100 images in 2.5 hours. Watch this video: https://www.youtube.com/watch?v=6cS9N4NqEIs (https://www.youtube.com/watch?v=6cS9N4NqEIs)
If I remember correctly there was a contributor here on the forum who had uploaded 3000 AI images in a month (100 per day).
-
Producing 100 images with AI takes just two or three hours?
do you mean to generate unfinished images,just prompts?yes,you can do even more in 2-3 hours,but the work is just started.
I just sent 10 AI images to Adobe and it took me 2 days!
I'm no expert, but I see that many succeed.
They may not be the best images in the world but it is possible to take 100 images in 2.5 hours. Watch this video: https://www.youtube.com/watch?v=6cS9N4NqEIs (https://www.youtube.com/watch?v=6cS9N4NqEIs)
If I remember correctly there was a contributor here on the forum who had uploaded 3000 AI images in a month (100 per day).
yes that's the mentality,that's why Adobe Stock ends up with a bunch of AI photos with lots of generation errors,because people just want to generate,expand(in a quick way)and upload.
I hope that the AI upload limits on Adobe Stock are limited to a maximum of 10,000 in a year,or even less,and that they clean up all these images with logos,or with bad generation that escape review.
I'm sure that even if they are approved in the beginning,will then be checked and deleted if the images have generation errors or too much similar,or other reasons.
however there are limits on the AI content you can upload to Adobe,so I don't see the point in all this rush.
is it right that there are limits,or this is no longer creating but madness.
I think that 10,000 AI contents maximum per account per year it would be better for everyone and also to have quality content in the Adobe Stock collection and also to not overload the reviewers and have shorter review times.
We all stand to gain if Adobe places a limit and communicates it officially,on the number of AI content that each contributor can have each year on sale.
-
I hope that the AI upload limits on Adobe Stock are limited to a maximum of 10,000 in a year,or even less,and that they clean up all these images with logos,or with bad generation that escape review.
I think that 10,000 AI contents maximum per account per year it would be better for everyone and also to have quality content in the Adobe Stock collection and also to not overload the reviewers and have shorter review times.
We all stand to gain if Adobe places a limit and communicates it officially,on the number of AI content that each contributor can have each year on sale.
It will not work as people are so clever that they will create multiple accounts to increase their upload limit.
Everyone wants to earn and they don't care how they earn or what spamming they are doing.
Agencies needs to take action to control these stuffs to promote more authentic work and real artists.
Sometimes I feel there is a huge requirement for a Microstock Organization and all RF and RM companies should come under them. Through this they can control pricing and authenticity both. Moreover can also organize a annual microstock meet to reward and also motivate people to work for such field.
-
I find it all strange anyway.
People are paying money to platforms on which they can create images that cannot be protected by copyright.
These platforms themselves earn money by infringing copyright without paying any compensation to those whose copyright they infringe.
It's all absolutely absurd!
It will become even more absurd when those who offer such images commercially and have also spent money on creating the images find that these images are copied exactly 1:1, because there is no copyright protection.
The whole construct is absolutely sick! It will lead to money being shifted around once again. The winners will be a handful of companies that offer AI technology. The losers will be those who have made a living from the creation of intellectual, literary, musical or pictorial works.
We will have even more monopolists, such as Google, who will then have global market power and dictate prices.
-
I hope that the AI upload limits on Adobe Stock are limited to a maximum of 10,000 in a year,or even less,and that they clean up all these images with logos,or with bad generation that escape review.
I think that 10,000 AI contents maximum per account per year it would be better for everyone and also to have quality content in the Adobe Stock collection and also to not overload the reviewers and have shorter review times.
We all stand to gain if Adobe places a limit and communicates it officially,on the number of AI content that each contributor can have each year on sale.
It will not work as people are so clever that they will create multiple accounts to increase their upload limit.
Everyone wants to earn and they don't care how they earn or what spamming they are doing.
Agencies needs to take action to control these stuffs to promote more authentic work and real artists.
Sometimes I feel there is a huge requirement for a Microstock Organization and all RF and RM companies should come under them. Through this they can control pricing and authenticity both. Moreover can also organize a annual microstock meet to reward and also motivate people to work for such field.
I agree 100%,but that's another question,those who create multiple accounts to circumvent the rules pay the consequences,you can be sure of that,the fact is that sometimes it takes a while to catch them.
I firmly believe that to reduce AI content spam,and don't create this crazy rush of 100 contents in 2 hours,favor review,favor real content,lighten the load,increase the quality produced... for many reasons I firmly believe that Adobe should place a limit and announce it officially.
let this limit be 1000 per month or whatever they want,but there must be an officially communicated limit, so that people get their act together, and try to create more original and quality things instead of starting an "assembly line"
I'm sure Adobe has already thought about it.
Mat,if you read me,try to raise this issue with the team if is possible and if you can.
See if it's possible to have an officially announced limit of AI content per month or per year per account,I think 1000/1500 per month is more than enough,otherwise we'll all go crazy here,including you! :D
-
I agree 100%,but that's another question,those who create multiple accounts to circumvent the rules pay the consequences,you can be sure of that,the fact is that sometimes it takes a while to catch them.
I firmly believe that to reduce AI content spam,and don't create this crazy rush of 100 contents in 2 hours,favor review,favor real content,lighten the load,increase the quality produced... for many reasons I firmly believe that Adobe should place a limit and announce it officially.
let this limit be 1000 per month or whatever they want,but there must be an officially communicated limit, so that people get their act together, and try to create more original and quality things instead of starting an "assembly line"
I'm sure Adobe has already thought about it.
Mat,if you read me,try to raise this issue with the team if is possible and if you can.
See if it's possible to have an officially announced limit of AI content per month or per year per account,I think 1000/1500 per month is more than enough,otherwise we'll all go crazy here,including you! :D
Limits on new accounts are ok, limits on legacy accounts from contributors who know what they're doing - nah
Why implement this limit only on genAI? You can take your smartphone, take literally 10,000 photos in an hour, and send them all for review. That's a lot more than you can create with AI. Yet, nobody is advocating for upload limits for traditional photos.
It's the job of the reviewer to reject all images that don't satisfy criteria. You can't expect the contributors to "self censor" when it's not in their best interest. If they have a 1% more chance of selling by uploading an extra 10 images, they will do it. If someone can generate 10,000 good images per month - this means, good subject, inpainted faces and hands, upscaled with SD ultimate upscale - then great, let them! The issue doesn't come from those contributors. It comes from those who buy a Midjourney subscription, generate 1000 images, upload them all after upscaling in Topaz, and they pass review. The underlined part is the issue. Adobe doesn't have enough skilled reviewers to distinguish between good and bad upscales. They talk a lot about "educating contributors" but they should educate their reviewers.
If it were up to me, I'd delete over 90% of their current genAI collection - it's crap, mostly midjourney output upscaled in Topaz or something similar, no realistic skin textures, and it all looks samey.
There are a few contributors (and I count myself among them not because of an inflated ego but because of the skills I needed to learn to get where I am) who submit proper content, proper technically, proper theme, proper upscale and inpaint. With consumer hardware, it would be very hard to reach 10k images per month like that. Not impossible, but hard, and your machine would need to be generating and upscaling almost 24/7.
Once Adobe recognizes that it's not those contributors who are the problems, but the mass from a certain low-COL geographic region (no racism, just go on youtube and search for adobe stock ai earnings and see who's uploading) who spam their system with their crappy midjourney upscales, and the reviewers who, for whatever reason, allow this type of content in the adobe stock library.
Me personally - I am waiting for an AI-only "high end" stock agency, where such stuff will not be allowed. I've given up hope that Adobe will ever clean shop.
-
Another thing to consider is that Adobe Stock subscribers will have 500 credits to generate,this is not good news for me,it would have been better if the credits were only available to Adobe software subscribers,or less credits for Adobe Stock customers,500 for a month are more than enough for a customer I think.
in any case we'll see,in a year or less,we'll all have clearer answers,I believe and hope that Adobe has considered all this,we'll see! :)
But surely Shutterstock isn't accepting AI-images ..?
But Adobe users also need these credits for generative fill or other generative uses within Photoshop.
https://helpx.adobe.com/firefly/using/generative-credits.html
So I wouldn't worry too much about losing sales to image generation. As we all know that is still very time consuming.
-
I have no doubt that at the rate artificial intelligence is advancing, stock photography has a maximum of two years left to live. Right now the smart thing to do is to sell all the lenses, equipment and cameras on ebay before it is too late and they become worthless junk.
I don't think it will be quite that bad.
Cameras won't die out. Hardly anyone will have their travel or wedding pictures recreated with AI, for example ;).
Of course portrait-, news- and some site specific photo will not die. But generic photographs - and most illustrations! - will die.
But it will of course hit some photographers hard. For example, I wonder whether real food photography still makes sense with all the effort involved.
As my portfolio mainly consists of real locations and editorials, I don't see myself at risk here for the time being and will keep my camera.
-
I have no doubt that at the rate artificial intelligence is advancing, stock photography has a maximum of two years left to live. Right now the smart thing to do is to sell all the lenses, equipment and cameras on ebay before it is too late and they become worthless junk.
I haven't sent any AI images, and don't think I ever will.
I get help from AI for my post-production, but I'm not interested in generative AI.
I agree. But anyway there is lots of complicated post production involved in creating good AI-images.
It's boring, and the images are all the same, all with the same style.
Producing 100 images with AI takes just two or three hours, so photography and illustration categories will soon be saturated with millions of images uploaded in no time.
I think eventually, people will get bored, and the majority of AI users will create their own images.
I hope in the future, there will always be a niche of customers who prefer real images captured with authentic landscapes, models, and objects. And I continue to work for that niche.
We'll see in a few years.
-
Producing 100 images with AI takes just two or three hours?
do you mean to generate unfinished images,just prompts?yes,you can do even more in 2-3 hours,but the work is just started.
I just sent 10 AI images to Adobe and it took me 2 days!
"this is no longer creating but madness" .. right!
I'm no expert, but I see that many succeed.
They may not be the best images in the world but it is possible to take 100 images in 2.5 hours. Watch this video: https://www.youtube.com/watch?v=6cS9N4NqEIs (https://www.youtube.com/watch?v=6cS9N4NqEIs)
If I remember correctly there was a contributor here on the forum who had uploaded 3000 AI images in a month (100 per day).
yes that's the mentality,that's why Adobe Stock ends up with a bunch of AI photos with lots of generation errors,because people just want to generate,expand(in a quick way)and upload.
I hope that the AI upload limits on Adobe Stock are limited to a maximum of 10,000 in a year,or even less,and that they clean up all these images with logos,or with bad generation that escape review.
I'm sure that even if they are approved in the beginning,will then be checked and deleted if the images have generation errors or too much similar,or other reasons.
however there are limits on the AI content you can upload to Adobe,so I don't see the point in all this rush.
is it right that there are limits,or this is no longer creating but madness.
I think that 10,000 AI contents maximum per account per year it would be better for everyone and also to have quality content in the Adobe Stock collection and also to not overload the reviewers and have shorter review times.
We all stand to gain if Adobe places a limit and communicates it officially,on the number of AI content that each contributor can have each year on sale.
-
I agree 100%,but that's another question,those who create multiple accounts to circumvent the rules pay the consequences,you can be sure of that,the fact is that sometimes it takes a while to catch them.
I firmly believe that to reduce AI content spam,and don't create this crazy rush of 100 contents in 2 hours,favor review,favor real content,lighten the load,increase the quality produced... for many reasons I firmly believe that Adobe should place a limit and announce it officially.
let this limit be 1000 per month or whatever they want,but there must be an officially communicated limit, so that people get their act together, and try to create more original and quality things instead of starting an "assembly line"
I'm sure Adobe has already thought about it.
Mat,if you read me,try to raise this issue with the team if is possible and if you can.
See if it's possible to have an officially announced limit of AI content per month or per year per account,I think 1000/1500 per month is more than enough,otherwise we'll all go crazy here,including you! :D
Limits on new accounts are ok, limits on legacy accounts from contributors who know what they're doing - nah
Why implement this limit only on genAI? You can take your smartphone, take literally 10,000 photos in an hour, and send them all for review. That's a lot more than you can create with AI. Yet, nobody is advocating for upload limits for traditional photos.
It's the job of the reviewer to reject all images that don't satisfy criteria. You can't expect the contributors to "self censor" when it's not in their best interest. If they have a 1% more chance of selling by uploading an extra 10 images, they will do it. If someone can generate 10,000 good images per month - this means, good subject, inpainted faces and hands, upscaled with SD ultimate upscale - then great, let them! The issue doesn't come from those contributors. It comes from those who buy a Midjourney subscription, generate 1000 images, upload them all after upscaling in Topaz, and they pass review. The underlined part is the issue. Adobe doesn't have enough skilled reviewers to distinguish between good and bad upscales. They talk a lot about "educating contributors" but they should educate their reviewers.
If it were up to me, I'd delete over 90% of their current genAI collection - it's crap, mostly midjourney output upscaled in Topaz or something similar, no realistic skin textures, and it all looks samey.
There are a few contributors (and I count myself among them not because of an inflated ego but because of the skills I needed to learn to get where I am) who submit proper content, proper technically, proper theme, proper upscale and inpaint. With consumer hardware, it would be very hard to reach 10k images per month like that. Not impossible, but hard, and your machine would need to be generating and upscaling almost 24/7.
Once Adobe recognizes that it's not those contributors who are the problems, but the mass from a certain low-COL geographic region (no racism, just go on youtube and search for adobe stock ai earnings and see who's uploading) who spam their system with their crappy midjourney upscales, and the reviewers who, for whatever reason, allow this type of content in the adobe stock library.
Me personally - I am waiting for an AI-only "high end" stock agency, where such stuff will not be allowed. I've given up hope that Adobe will ever clean shop.
We can hope that Adobe drowns in customer complains ...!
-
I agree 100%,but that's another question,those who create multiple accounts to circumvent the rules pay the consequences,you can be sure of that,the fact is that sometimes it takes a while to catch them.
Again, spammers are not here to build a long term good portfolio or any reputation. They are here for a short term game and are finding this as easy money option.
-
I agree 100%,but that's another question,those who create multiple accounts to circumvent the rules pay the consequences,you can be sure of that,the fact is that sometimes it takes a while to catch them.
I firmly believe that to reduce AI content spam,and don't create this crazy rush of 100 contents in 2 hours,favor review,favor real content,lighten the load,increase the quality produced... for many reasons I firmly believe that Adobe should place a limit and announce it officially.
let this limit be 1000 per month or whatever they want,but there must be an officially communicated limit, so that people get their act together, and try to create more original and quality things instead of starting an "assembly line"
I'm sure Adobe has already thought about it.
Mat,if you read me,try to raise this issue with the team if is possible and if you can.
See if it's possible to have an officially announced limit of AI content per month or per year per account,I think 1000/1500 per month is more than enough,otherwise we'll all go crazy here,including you! :D
Limits on new accounts are ok, limits on legacy accounts from contributors who know what they're doing - nah
Why implement this limit only on genAI? You can take your smartphone, take literally 10,000 photos in an hour, and send them all for review. That's a lot more than you can create with AI. Yet, nobody is advocating for upload limits for traditional photos.
It's the job of the reviewer to reject all images that don't satisfy criteria. You can't expect the contributors to "self censor" when it's not in their best interest. If they have a 1% more chance of selling by uploading an extra 10 images, they will do it. If someone can generate 10,000 good images per month - this means, good subject, inpainted faces and hands, upscaled with SD ultimate upscale - then great, let them! The issue doesn't come from those contributors. It comes from those who buy a Midjourney subscription, generate 1000 images, upload them all after upscaling in Topaz, and they pass review. The underlined part is the issue. Adobe doesn't have enough skilled reviewers to distinguish between good and bad upscales. They talk a lot about "educating contributors" but they should educate their reviewers.
If it were up to me, I'd delete over 90% of their current genAI collection - it's crap, mostly midjourney output upscaled in Topaz or something similar, no realistic skin textures, and it all looks samey.
There are a few contributors (and I count myself among them not because of an inflated ego but because of the skills I needed to learn to get where I am) who submit proper content, proper technically, proper theme, proper upscale and inpaint. With consumer hardware, it would be very hard to reach 10k images per month like that. Not impossible, but hard, and your machine would need to be generating and upscaling almost 24/7.
Once Adobe recognizes that it's not those contributors who are the problems, but the mass from a certain low-COL geographic region (no racism, just go on youtube and search for adobe stock ai earnings and see who's uploading) who spam their system with their crappy midjourney upscales, and the reviewers who, for whatever reason, allow this type of content in the adobe stock library.
Me personally - I am waiting for an AI-only "high end" stock agency, where such stuff will not be allowed. I've given up hope that Adobe will ever clean shop.
but it's a completely different thing, with AI you can create 1000 completely different contents in a couple of days if you don't pay attention to details and do everything quickly.
I'm sure that having an official limit means that Adobe's review system can be more efficient,and it's better for everyone and for many reasons.
-
the limit for AI content already exists,we just don't know what it is,and I think it's better to know for many reasons.
in the contributor terms:
"We don’t permit multiple account creations to submit similar or identical content to inflate sales,or to bypass upload limits for generative AI content".
as you can see,if this rule exists,it means that there is already a limit for AI content.
if it is not possible to know this limit ok,but I think it is better for everyone to know it.
-
the limit for AI content already exists,we just don't know what it is,and I think it's better to know for many reasons.
in the contributor terms:
"We don’t permit multiple account creations to submit similar or identical content to inflate sales,or to bypass upload limits for generative AI content".
as you can see,if this rule exists,it means that there is already a limit for AI content.
if it is not possible to know this limit ok,but I think it is better for everyone to know it.
They had the same rules before AI, most agencies, including those that don't take AI, have the same rules, one account allowed. Upload limits are real, your theory that there are acceptance limits is just your guess.
-
the limit for AI content already exists,we just don't know what it is,and I think it's better to know for many reasons.
in the contributor terms:
"We don’t permit multiple account creations to submit similar or identical content to inflate sales,or to bypass upload limits for generative AI content".
as you can see,if this rule exists,it means that there is already a limit for AI content.
if it is not possible to know this limit ok,but I think it is better for everyone to know it.
They had the same rules before AI, most agencies, including those that don't take AI, have the same rules, one account allowed. Upload limits are real, your theory that there are acceptance limits is just your guess.
is not so,because it is clearly written that it is not allowed to create multiple accounts to bypass upload limits for generative AI content.
I'll ask you a more direct question:"Wouldn't it be better for you to know that there is for example a limit of 1000 AI contents that can be accepted per account per month?"
for me it would be better,so i don't send 2000 AI contents just for finding myself with 1000 rejections,it would just be a waste of time.
it would be a deterrent for those who want to create multiple accounts,because they would know from the start that there is a limit of 1000 AI contents per month.
it would be better for those who want to send 10,000 AI contents per month for sale without even looking at them,so they don't.
it would be better for not overload the review.
it would be better to not completely cover real contents.
There are so many reasons why it would be better that I'm starting to forget them! :D
the point is that there may be other reasons why this type of information needs to remain confidential.
I support transparency,and Adobe Stock is an agency that has always supported transparency,even if at times it has found itself making difficult and necessary choices which too often have not been understood,here because for this reason that I take the liberty of raising this issue.
I believe it is better for us contributors to know a defined limit of AI content acceptable for sale,and you should too!
-
For those who haven't been following the collection sizes, Adobe Stock's genAI collection is almost at 23 million; Freepik's is over 30 million; Dreamstime's is a hair over 7 million.
Shutterstock's is such dreck it doesn't matter, but it's approaching 1.5 million
The "oops" images continue to be everywhere...
(https://img.freepik.com/premium-photo/photo-two-asian-student-kid-girl-boy-schoolchildren-brother-sister-smile-happy_584200-5293.jpg) (https://www.freepik.com/premium-ai-image/photo-two-asian-student-kid-girl-boy-schoolchildren-brother-sister-smile-happy_84531340.htm#fromView=recent&page=3&position=6&track=®ularType=ai)
(https://t3.ftcdn.net/jpg/06/57/07/88/240_F_657078853_2sONF4dTMuXAmbCXUDVRvfYLTdSSGG2h.jpg) (https://stock.adobe.com/images/hyper-realistic-bananas-in-cozy-market-setting/657078853) (https://t3.ftcdn.net/jpg/06/53/99/48/240_F_653994861_O2UHk4Iy30hDGSljGUUE4AMUIMCZAuFL.jpg) (https://stock.adobe.com/images/roasted-chicken-with-salad-and-other-food-items/653994861)
At least no one will be fighting over the drumstick on Thanksgiving 8)
-
the limit for AI content already exists,we just don't know what it is,and I think it's better to know for many reasons.
in the contributor terms:
"We don’t permit multiple account creations to submit similar or identical content to inflate sales,or to bypass upload limits for generative AI content".
as you can see,if this rule exists,it means that there is already a limit for AI content.
if it is not possible to know this limit ok,but I think it is better for everyone to know it.
They had the same rules before AI, most agencies, including those that don't take AI, have the same rules, one account allowed. Upload limits are real, your theory that there are acceptance limits is just your guess.
is not so,because it is clearly written that it is not allowed to create multiple accounts to bypass upload limits for generative AI content.
I'll ask you a more direct question:"Wouldn't it be better for you to know that there is for example a limit of 1000 AI contents that can be accepted per account per month?"
The two don't mean the same thing, and your assumption is that there is a secret acceptance limit, which is confidential. But there's nothing to say there really is any acceptance limit, just a submission limit.
I'll ask a simple question. If there's a secret acceptance limit, how do you know that's a fact? Or is it just something that you suppose is hypothetically possible? As Yada wrote, there has been a one account limit, since the start, for all agencies. This isn't just because of AI or some classified acceptance cap.
The same rule has always been in place to prevent people from exceeding the upload limits and from preventing them from gaming the collection, possibly improving sales, with multiple versions, appearing in multiple accounts. Before AI.
-
At the beginning of October, Adobe Stock was adding around 900k genAI images per week.
From October 16th for the next three weeks, it was just over 1 million each week.
From Nov 6-13, it was almost 2 million!! 1,930,975.
I very much doubt that Adobe Stock is adding buyers at anything like the rate that it's adding genAI images. Or that existing buyers are suddenly going to be buying a significantly larger number of items. I doubt the AI gold-rush enthusiasts have thought too much about where this is all going, but even if this was all novel content I think there'd be a supply & demand imbalance.
Freepik's collection shrank between last Monday & today - from 30.08m down to 29.63m. It had grown from 28.7 to 30.08 the previous week, so I assume they were doing some cleanup?
-
At the beginning of October, Adobe Stock was adding around 900k genAI images per week.
From October 16th for the next three weeks, it was just over 1 million each week.
From Nov 6-13, it was almost 2 million!! 1,930,975.
I very much doubt that Adobe Stock is adding buyers at anything like the rate that it's adding genAI images. Or that existing buyers are suddenly going to be buying a significantly larger number of items. I doubt the AI gold-rush enthusiasts have thought too much about where this is all going, but even if this was all novel content I think there'd be a supply & demand imbalance.
Freepik's collection shrank between last Monday & today - from 30.08m down to 29.63m. It had grown from 28.7 to 30.08 the previous week, so I assume they were doing some cleanup?
And out of those 2 million added, my feeling is that 10% at most should have passed review. I looked at some of them in 1:1 and it's really funny what Adobe reviewers think is acceptable.
-
At least no one will be fighting over the drumstick on Thanksgiving 8)
It's quite funny, actually, because the authenticity of the Ki images is in many cases lousy, and this is not sorted out by either the creators or the image checkers. A worst case scenario.
With regard to the "Christmas goose" - perhaps the world should now be prepared for artificial meat and artificial meat products... 8)
-
But to be balanced…a lot of the ai content coming in is incredibly good.
The low quality work will get cleaned out by the algos.
Yes, they could be stricter, because it feels unfair to the photos, but a lot of the new stuff is fantastic and a real advantage for buyers compared to all the other agencies.
-
It's interesting to think of this shift as a parallel to the many other technological shifts that have happened in the past. When Microsoft Excel/PCs came out it didn't put accountants out of business, but it surely changed the way they work forever.
-
30 Million+
-
Producing 100 images with AI takes just two or three hours?
do you mean to generate unfinished images,just prompts?yes,you can do even more in 2-3 hours,but the work is just started.
I just sent 10 AI images to Adobe and it took me 2 days!
I generated them with AI yes,but then I completely renovated,improved,added particular effects...it takes forever!
photographers and videographers will always be necessary,AI changes the game but cannot replace real content.
...
totally agree - those who keep citing mass quantities of similar images (as if it's new challenge to 'real' artists) seem to have forgotten the thousands of silly images that have been approved over the years ( tomato slices, )
-
... their upload limit.
...
Agencies needs to take action to control these stuffs to promote more authentic work and real artists.
Sometimes I feel there is a huge requirement for a Microstock Organization and all RF and RM companies should come under them. Through this they can control pricing and authenticity both. Moreover can also organize a annual microstock meet to reward and also motivate people to work for such field.
luckily most of us don't live in a country that would be able to dictate what private companies must do.
-- who would control pricing?
-- how would you force membership?
-- how would this work when agencies are global?
-- who'd decide what is acceptable or authentic?
-
...
At least no one will be fighting over the drumstick on Thanksgiving 8)
but they should give credit to John Madden who invented the multi-drumstick turkey for their NFL Thanksgiving broadcasts
-
At the beginning of October, Adobe Stock was adding around 900k genAI images per week.
From October 16th for the next three weeks, it was just over 1 million each week.
From Nov 6-13, it was almost 2 million!! 1,930,975.
...
for comparison, what were the numbers for non-AI? that would help to see how much AI is affecting the total
-
i don't know about Adobe, but back when Shutterstock was publishing these stats it was around 2 million files a week over everything (photo, illustration, video).
Adobe is probably pickier, but with the long inspection times their submission volume must have more than doubled
-
Not sure what the Adobe review waiting time is, but I have a file waiting for inspection 28 days.....frustrating.
-
As always the question comes down to what we should be doing instead. And unfortunately most of us don't have an alternative figured out yet. So it of habit and compulsion we'll keep sending to stock
-
350k a day
-
Uploading a photo was throwing a drop in a pool. Now, throwing a drop in an ocean... ::)
-
And here we are, still making money.
We survived unsplash...what could be worse than free images?
The majority of content uploaded is just duplicates of duplicates.
Once you dig into a genre you see all the glaring holes in the collections.
Huge amounts of content are missing, but most uploaders just sort a search by most downloaded and copy the top 20 files. Again and again.
Give the customers something even just a little bit different and they will be grateful and buy.
-
We have now reached over 33 million AI photos and illustrations on adobe stock.
Of these, less than 3 million have been sold, which corresponds to a share of just 7%.
The majority is therefore literally data garbage that has no buyers.
@Cobalt
I think that there will no longer be a market for generic motifs (people, life style, interior, etc.) and that this will be completely replaced by the stock providers' own AI models.
Where there is still some market potential are only niches, where AI can't produce high detailed images without errors.
Contributors like https://stock.adobe.com/de/contributor/205024019/ipopba are now melking for short future the current demanded niches.
Some successful contributors specialized on generic motifs have already switched to the video market in time.
-
We have now reached over 33 million AI photos and illustrations on adobe stock.
Of these, less than 3 million have been sold, which corresponds to a share of just 7%.
The majority is therefore literally data garbage that has no buyers.
@Cobalt
I think that there will no longer be a market for generic motifs (people, life style, interior, etc.) and that this will be completely replaced by the stock providers' own AI models.
Where there is still some market potential are only niches, where AI can't produce high detailed images without errors.
Contributors like https://stock.adobe.com/de/contributor/205024019/ipopba are now melking for short future the current demanded niches.
Some successful contributors specialized on generic motifs have already switched to the video market in time.
Just for comparison, what percentage of real photos have been sold from the entire collection?
AI in so much alike and so many that more of the same helps no one. Not us or the agency. When does Adobe say, enough is enough? At 66 Million or when the new images change nothing for the volume of downloads? There has to be a wall.
-
Just for comparison, what percentage of real photos have been sold from the entire collection?
AI in so much alike and so many that more of the same helps no one. Not us or the agency. When does Adobe say, enough is enough? At 66 Million or when the new images change nothing for the volume of downloads? There has to be a wall.
According to the Adobe search engine in total around 194 million of the 242 million classic photos and illustrations were never sold.
This corresponds to a share of almost exactly 80%.
Conversely, 20% were sold at least once.
In my view, this is a plausible result and can often be observed as the Pareto principle in markets with very unequal concentration distributions.
It is possible that the AI images will sell better in the future after there has been a market shakeout of the bad skilled contributors.
But I'm also very curious to see whether Adobe will close the floodgates for AI content this year.
I can't imagine that they can manage 2 to 3 million new images every few weeks in the long term.
I also fear that the ranking and keyword order will suffer damage. There are so many images either with the wrong tags or in the wrong categories.
All in all I am surprised that nobody tries to individualize the Midjourney image look with own filters.
It's like a collective frenzy at the moment, who can generate more content.
-
Just for comparison, what percentage of real photos have been sold from the entire collection?
AI in so much alike and so many that more of the same helps no one. Not us or the agency. When does Adobe say, enough is enough? At 66 Million or when the new images change nothing for the volume of downloads? There has to be a wall.
According to the Adobe search engine in total around 194 million of the 242 million classic photos and illustrations were never sold.
This corresponds to a share of almost exactly 80%.
Conversely, 20% were sold at least once.
In my view, this is a plausible result and can often be observed as the Pareto principle in markets with very unequal concentration distributions.
It is possible that the AI images will sell better in the future after there has been a market shakeout of the bad skilled contributors.
But I'm also very curious to see whether Adobe will close the floodgates for AI content this year.
I can't imagine that they can manage 2 to 3 million new images every few weeks in the long term.
I also fear that the ranking and keyword order will suffer damage. There are so many images either with the wrong tags or in the wrong categories.
All in all I am surprised that nobody tries to individualize the Midjourney image look with own filters.
It's like a collective frenzy at the moment, who can generate more content.
20% of the entire collection with at least one sale is an amazing number. Although people here, have listed their stats for sales, and are higher than that. I'm lower.
Keep in mind that a number of agencies and other places, offer AI image creation, directly for consumers. Anyone can go to many of those and make their own images, get just what they want, without artists, so they don't need to buy one from a stock site.
I have to laugh that you worry about rank, tags and keywords. I think we've been writing the same thing about stock sites since I started and that was 17 years ago. Spam keywords, deceptive tags, just plain wrong locations and tags, and never in the whole time of guessing how image ranks work, has anyone come up with anything that shows a benefit to keyword spam, or incorrect information that leads to more sales or better rank. In fact, if the system works at all, bad data would harm the rank of images with spam or inconsistencies.
Use Alamy clicks and views, and zooms and sales, as a starting point how other agencies are likely to rank images. Too many negative results and the image drops in rank. I don't think any of this is as simple as views and sales, it would be illogical to think and agency is that thoughtless to have some simple rudimentary system.
And then there are the "secret" reviewers rank, when an image is uploaded. Might be AI now, but you could upload 20 of the same image, and one of them will be ranked much higher, show pages higher, and come up more often in a search. Wouldn't they all be equal? Uploads have a rank, before they are live on the site. Spam, bad work, marginal content and junk files, won't have a higher than average starting rank.
-
Also remember when comparing AI to Photo sales that all the AI images are comparatively very very new.
-
Not sure how long there have been Premium AI images - $249.99 for large, $119.99 for small.Why on earth would a buyer pay that much for AI generated stuff? None of the typical reasons for the premium collection apply to genAI images. Nothing at all wrong with this image BTW, other than the price
(https://t4.ftcdn.net/jpg/07/07/32/25/240_F_707322528_ihLBnRTzu9EsqlAvtom4R737EjNKZqnx.jpg) (https://stock.adobe.com/images/solo-suitcase-on-airport-conveyor-belt-at-night-generative-ai/707322528)
-
This discussion is very reminiscent of comparing the film image to the first steps of digital. The difference is that the quality in AI images develops 10 times faster and here we don't just have pixels, but compositions, light and everything else.
-
Not sure how long there have been Premium AI images - $249.99 for large, $119.99 for small.Why on earth would a buyer pay that much for AI generated stuff? None of the typical reasons for the premium collection apply to genAI images. Nothing at all wrong with this image BTW, other than the price
(https://t4.ftcdn.net/jpg/07/07/32/25/240_F_707322528_ihLBnRTzu9EsqlAvtom4R737EjNKZqnx.jpg) (https://stock.adobe.com/images/solo-suitcase-on-airport-conveyor-belt-at-night-generative-ai/707322528)
Premium AI images? There's something that just seems all wrong about that.
-
Why did buyers pay hundreds of dollars for my iphone food snapshots when eyeem was a premium collection on getty?
Same reasons apply to ai.
-
I do stock as I enjoy photography and fresh air, can't see any enjoyment from doing AI.