MicrostockGroup
Agency Based Discussion => iStockPhoto.com => Topic started by: Brasilnut on January 05, 2025, 09:39
-
I've been using Topaz AI for noise reduction / sharpening for more than a year now and super pleased. However, I've noticed that iStockphoto have rejected whole batches of commercial and editorials. See screenshot below.
I went to find out why and from November 2024 they changed their policy. See link here:
https://contributors.gettyimages.com/article/10847
In summary:
If you want to make significant changes to your content, either separately or in addition to work you’ve done under the Retouching Requirements, follow these rules (“Modification”):
Do not apply Modification to your content with generative AI tools (only use traditional, non-generative AI tools).
Do not add location keywords, titles or descriptions more specific than Region/State/County if you modify recognizable, named, or famous locations.
Do not alter a model’s body shape to make them look thinner or larger than they are in real life. Find out more.
NOTE: Under these Modification Requirements, without using generative AI tools, you can:
Retouch more than 10% of the image’s total pixels.
Add new elements you own the copyright for, including creating composites.
Retouch your models more extensively.
Had no issues at any other agency. Frustrating, might just stop uploading to iStock anyway for what they pay and now further hurdles of creating two separate batches.
-
Maybe strip the meta or exif data before submitting.
-
Maybe strip the meta or exif data before submitting.
Removing the corresponding metadata (Exif tag ‘Software used’) could work.
Here is a screenshot of our own Zoom software where you can see that Topaz leaves an entry that is possibly recognised by IStock and is considered AI-generated.
In fact, this is technically correct, because the AI system is very similar to a Ki image generator. In the end, not a single original pixel of your photo is retained - the image is generated block by block in a different resolution/de-noised/business.
Strictly speaking, it is no longer your image, but a completely generated image to which you no longer have the rights.
If you want to avoid such problems, you would have to avoid scaling/de-noising/sharpening programmes with generative AI and rely on other algorithms.
Examples of such scaling programmes with ‘non-generative AIs’ would be Zoom #2 professional or Photozoom #8 and in the area of denoising, Neat Image or Denoise #5 professional come to mind.
-
Keeping "-topaz-sharpen" in the filename probably isn't wise either.
-
For normal photos I process with an old ps elements version that doesn't have modern ai tools.
I hope that version stays available for a long time.
But next generation cameras will be using ai integrated into their software.
What happens then?
-
But next generation cameras will be using ai integrated into their software.
What happens then?
AI will be everywhere. The agencies are fighting a losing battle. I'm sure iStock and others are already flooded with AI-modified videos and images without them even knowing it.
-
In fact, this is technically correct, because the AI system is very similar to a Ki image generator. In the end, not a single original pixel of your photo is retained - the image is generated block by block in a different resolution/de-noised/business.
Strictly speaking, it is no longer your image, but a completely generated image to which you no longer have the rights.
Using an AI tool for sharpening certainly does not alter a picture in a way that you lose copyright, strictly speaking or otherwise.
That no pixel of the original image may be retained is hardly an argument, because that if true of a plethora of filters or other modification.
If you apply a filter to a photo, so that it looks like an impressionist painting with Photoshop, no pixel is the same either. Even if you just make a picture darker or brighter or change the colour temperature, no pixel may be exactly the same afterwards, if you change it enough. But you still have the copyright.
-
Using an AI tool for sharpening certainly does not alter a picture in a way that you lose copyright, strictly speaking or otherwise.
That no pixel of the original image may be retained is hardly an argument, because that if true of a plethora of filters or other modification.
If you apply a filter to a photo, so that it looks like an impressionist painting with Photoshop, no pixel is the same either. Even if you just make a picture darker or brighter or change the colour temperature, no pixel may be exactly the same afterwards, if you change it enough. But you still have the copyright.
The rights problem (which has not yet been finally clarified) also arises from the question of where the training data comes from. As soon as an image has been processed with a generative AI algorithm and not all of the image rights (without exception) of the training data are held by the creator of the software, this is a potential problem.
If a purely mathematical algorithm is used, no training data is required and the question of rights does not even arise.
I don't want to argue against you here, I just want to sensitise users to the problem. As I create image editing software myself (also with AI algorithms) and sell it, I deal with this problem almost every day, always ensuring that the rights issue is fully addressed for my users :)
-
I've been using Topaz AI for noise reduction / sharpening for more than a year now and super pleased. However, I've noticed that iStockphoto have rejected whole batches of commercial and editorials. See screenshot below.
I went to find out why and from November 2024 they changed their policy. See link here:
https://contributors.gettyimages.com/article/10847
In summary:
If you want to make significant changes to your content, either separately or in addition to work you’ve done under the Retouching Requirements, follow these rules (“Modification”):
Do not apply Modification to your content with generative AI tools (only use traditional, non-generative AI tools).
Do not add location keywords, titles or descriptions more specific than Region/State/County if you modify recognizable, named, or famous locations.
Do not alter a model’s body shape to make them look thinner or larger than they are in real life. Find out more.
NOTE: Under these Modification Requirements, without using generative AI tools, you can:
Retouch more than 10% of the image’s total pixels.
Add new elements you own the copyright for, including creating composites.
Retouch your models more extensively.
Had no issues at any other agency. Frustrating, might just stop uploading to iStock anyway for what they pay and now further hurdles of creating two separate batches.
As far as I am aware Topaz AI Denoise and Sharpen are supposed to be acceptable on IS, but if you are using some of Photoshop's tools, just check that AI has not been activated on them ... the latest version of CC turns them on automatically again.
-
For normal photos I process with an old ps elements version that doesn't have modern ai tools.
I hope that version stays available for a long time.
But next generation cameras will be using ai integrated into their software.
What happens then?
The latest version of Elements 2025, and the 2024 version is not available any longer, has a license that expires. "The license is for a full 3-year term, with no monthly or annual recurring subscription fees required. The license is non-renewing — it expires 3 years after redemption, at which time the Editor will no longer be accessible, but the Organiser will continue to be accessible indefinitely."
Yes, I use Elements too. I have Photoshop, Illustrator and Lightroom CC, but I still prefer Elements.
Maybe strip the meta or exif data before submitting.
Removing the corresponding metadata (Exif tag ‘Software used’) could work.
Here is a screenshot of our own Zoom software where you can see that Topaz leaves an entry that is possibly recognised by IStock and is considered AI-generated.
Right on the Zoom software.
Some of the AI software, and I don't know to what extent now, I haven't look recently, embeds text code in the images data, that is NOT in the EXIF or other metadata. Wiping the EXIF does not remove that. It's not that easy.
-
I've been using Topaz AI for noise reduction / sharpening for more than a year now and super pleased. However, I've noticed that iStockphoto have rejected whole batches of commercial and editorials. See screenshot below.
I went to find out why and from November 2024 they changed their policy. See link here:
https://contributors.gettyimages.com/article/10847
In summary:
If you want to make significant changes to your content, either separately or in addition to work you’ve done under the Retouching Requirements, follow these rules (“Modification”):
Do not apply Modification to your content with generative AI tools (only use traditional, non-generative AI tools).
Do not add location keywords, titles or descriptions more specific than Region/State/County if you modify recognizable, named, or famous locations.
Do not alter a model’s body shape to make them look thinner or larger than they are in real life. Find out more.
NOTE: Under these Modification Requirements, without using generative AI tools, you can:
Retouch more than 10% of the image’s total pixels.
Add new elements you own the copyright for, including creating composites.
Retouch your models more extensively.
Had no issues at any other agency. Frustrating, might just stop uploading to iStock anyway for what they pay and now further hurdles of creating two separate batches.
As far as I am aware Topaz AI Denoise and Sharpen are supposed to be acceptable on IS, but if you are using some of Photoshop's tools, just check that AI has not been activated on them ... the latest version of CC turns them on automatically again.
I'm using Topaz for noise reduction on most images, and occasionally use their sharpening tool. I haven't experienced any problems with Istock acceptance.
As far as I know, the potentially problematic Topaz tool is Gigapixel, as this uses AI to upscale.
-
Maybe strip the meta or exif data before submitting.
Not a good idea - another condition of the same policy update https://contributors.gettyimages.com/article/10847 (https://contributors.gettyimages.com/article/10847) says:
Do not remove or change metadata which content editing tools may embed in the file. Be sure to follow these rules and our latest Content Upload Requirements to avoid file rejections, account suspension, or account closure.
-
A few clarifications:
- Only used Topaz AI sharpening and/or noise reductions. I've disabled face recovery as some results are just weird.
- Didn't use Gigapixel upscaling
Next steps:
- Will test if by deleting the title will make a difference (I don't think so)
Curious theory:
- Apparently, human reviewers are paid by image/video and so it's easier for them to reject whole batches without much inspection. Perhaps they will begin adopting AI reviews as is now becoming the norm at other agencies (Depositphoto assets are accepted/rejected within seconds).
-
What happens if you do all the AI you want, flatten the image, save it as a JPG, open up an unrelated raw file (with no AI, no adjustments), drag the jpg onto that unrelated raw file and flatten it. Save it as whatever file name logic you use. Wouldn't that effectively strip out any evidence of AI in the exif data?
-
What happens if you do all the AI you want, flatten the image, save it as a JPG, open up an unrelated raw file (with no AI, no adjustments), drag the jpg onto that unrelated raw file and flatten it. Save it as whatever file name logic you use. Wouldn't that effectively strip out any evidence of AI in the exif data?
With tools like ExifTool, metadata records actions like flattening, saving, or combining files. Even if you remove metadata to hide these traces, that can make the file appear suspicious, as stripped metadata is often a red flag.
-
What happens if you do all the AI you want, flatten the image, save it as a JPG, open up an unrelated raw file (with no AI, no adjustments), drag the jpg onto that unrelated raw file and flatten it. Save it as whatever file name logic you use. Wouldn't that effectively strip out any evidence of AI in the exif data?
With tools like ExifTool, metadata records actions like flattening, saving, or combining files. Even if you remove metadata to hide these traces, that can make the file appear suspicious, as stripped metadata is often a red flag.
It would not be stripped. It would contain the meta data from the new raw file.
-
ChatGPT helped me a little bit:
Generally, metadata in image files (such as EXIF, IPTC, or XMP) does not record detailed editing actions like flattening layers, merging files, or specific save steps in Photoshop. Standard metadata fields typically include information such as:
Camera settings (exposure time, aperture, ISO, etc.)
Capture date and time
Geolocation (if enabled)
Copyright info
Software used (e.g., “Adobe Photoshop 2024”)
However, they usually do not store a history of how the file was altered (like which layers were flattened, in what sequence images were combined, or how many times the file was saved).
Where does Photoshop store editing actions?
History Panel (Temporary): Photoshop keeps a record of the most recent editing steps in the History panel, but this history is lost once you close the file or exit Photoshop (unless you specifically configure and save History Logs within Photoshop preferences, but that’s stored in a text/log file on your computer, not embedded in the image).
PSD or PSB Files (Layers & Metadata): Photoshop file formats (.psd, .psb) maintain layers and adjustments, but they still don’t embed each step you took to get there.
Can you embed editing steps at all?
In some workflows, you can save a layered PSD that preserves your edits and smart objects, but that’s not the same as a detailed action-by-action log.
History Logs can be turned on in Photoshop’s preferences (“History & Cache” settings), but those logs are external text files or appended as metadata in a very limited manner, and are generally not part of the standard EXIF/ITPC fields other software would read.
So, if you flatten, save, or merge files in Photoshop, that information typically does not appear in the exported image’s standard metadata. Instead, only a general reference to “Adobe Photoshop” and a timestamp for the last edit might be included.
If you paste an AI-generated image (Image A) over an unedited image (Image B) in Photoshop, flatten it, and save the result, Image B’s EXIF metadata will not automatically retain or include any metadata from Image A. Here’s why:
How Photoshop Handles Metadata During Editing
When Pasting or Combining Images:
Metadata from a pasted image (Image A) does not carry over to the final file. The act of pasting only transfers the visual content (pixels), not the metadata.
Flattening Layers:
When you flatten an image, Photoshop combines all layers into one. This operation discards the layer structure and any linked file references, but it doesn’t alter or add new metadata unless you explicitly modify it.
Metadata in the Saved File:
The saved image will retain the metadata of the base file (Image B) unless you change it manually in Photoshop or through external tools.
For example, the EXIF field might still indicate that Photoshop was used for editing, and it will record the last modification date. However, there will be no direct trace of Image A.
Stock Agency Concerns
Stock agencies that restrict AI-generated content typically rely on:
Visual Analysis: Identifying characteristics of AI-generated imagery.
Metadata Inspections: Checking for clues in the metadata that indicate AI tools or software were used.
By flattening and saving as one layer, the metadata from Image A will not be embedded in the final image unless:
The AI software explicitly watermarked the image or inserted detectable data into the visual content.
You intentionally added metadata indicating the use of AI tools.
Best Practices and Considerations:
If you're submitting images to stock agencies, ensure the final content complies with their policies. Flattening may "hide" metadata, but it doesn’t change the fact that the content was partially AI-generated.
Some stock agencies use AI detection tools that analyze pixel patterns and artifacts, which could still identify the image as AI-generated even if metadata is removed.
While metadata manipulation might work technically, it’s ethically better to adhere to agency guidelines and disclose the use of AI where required.
-
Firefly if you open an image in a text editor. This is not in the EXIF section.
(https://i.postimg.cc/bwDgPcFg/firefly-image-embedded-code.jpg)
C2PA
https://c2pa.org/
"The Coalition for Content Provenance and Authenticity (C2PA) addresses the prevalence of misleading information online through the development of technical standards for certifying the source and history (or provenance) of media content. C2PA is a Joint Development Foundation project, formed through an alliance between Adobe, Arm, Intel, Microsoft and Truepic."
-
However, ... if you paste an AI-generated image (Image A) over an unedited image (Image B) in Photoshop, flatten it, and save the result, Image B’s EXIF metadata will not retain or include any metadata from Image A. I think it will have NO data from image A, except the pixels copied. Basta. All the info, in the EXIF or out of it, will be the one Image B had from the beginning, except "edited in Photoshop" will be added almost for sure.
-
Istock has nowhere written that you can't improve the quality of images and videos with Topaz Ai tools, upscaling or smoother slow motion, etc. And I guess the same is true for other agencies if they want better quality images. These Ai tools also help a lot for animation graphic designers because they have less work to do with renders, etc.
Stock agencies are more interested in whether there are image elements from MidJourney etc. in the images they submit. Such Ai generators are of no interest to me for commercial use.
Even Dxo Lab boasts that it improves RAW image quality in Deep prime Ai using a model with billions of images.
Just does this company have licenses for billions of images and permission from the authors? The same is true of Topaz Ai.
How do we know if we own the copyright to our image after developing RAW Deep prime Ai images?
The problem is that we don't know if we have copyright on our images after using these Ai image enhancement tools, and whether the stock agencies protect us from these problems if it came out that these programs used non-legal AI models.
I guess it would have been worse if we had sold Ai-enhanced images to customers ourselves without the intermediaries of stock agencies.
In my opinion, as one bought these programs legally, these companies should be responsible for Ai tools and not us customers.
There is also the AI Act in the European Union and the world's first comprehensive legal regulation for artificial intelligence systems and models.
I also found an interesting other similar thread about Ai.
https://www.reddit.com/r/TopazLabs/comments/17ptl6p/topaz_photo_commercial_usage/?sort=new (https://www.reddit.com/r/TopazLabs/comments/17ptl6p/topaz_photo_commercial_usage/?sort=new)
-
Istock has nowhere written that you can't improve the quality of images and videos with Topaz Ai tools, upscaling or smoother slow motion, etc. And I guess the same is true for other agencies if they want better quality images. These Ai tools also help a lot for animation graphic designers because they have less work to do with renders, etc.
Stock agencies are more interested in whether there are image elements from MidJourney etc. in the images they submit. Such Ai generators are of no interest to me for commercial use.
Sure makes sense, doesn't that?
However, ... if you paste an AI-generated image (Image A) over an unedited image (Image B) in Photoshop, flatten it, and save the result, Image B’s EXIF metadata will not retain or include any metadata from Image A. I think it will have NO data from image A, except the pixels copied. Basta. All the info, in the EXIF or out of it, will be the one Image B had from the beginning, except "edited in Photoshop" will be added almost for sure.
It's NOT in the EXIF or usual metadata. From what I've found the code is embedded in the image itself.
"When an image is created or edited using C2PA-compatible software, a set of cryptographically signed metadata is embedded directly into the image file. Content: This metadata includes information such as the creation time, the software used, and any edits made to the image."
Copy and paste into a different or new image, does not erase this tracking.
-
"If" you only used "Topaz AI Denoise" and or "Topaz AI sharpen" then IS has said that this is acceptable. "If" you think your images were rejected incorrectly then put in a ticket. I have not heard of anyone being rejected for those tools as they dont use a "data set" they use information from within the file.
But "if" you use Photoshop CC, and use tools like "content aware" corrections, a lot of them now have AI enabled and you need to turn it off. User beware in other words! Even though they say the minimal use is acceptable it usually scores a rejection.
Topaz AI Gigapixel is not allowed at all, but resizing up is frowned upon anyway.
Metadata removal will also get you a rejection.
As Uncle Pete said the information is embedded within the file not in the metadata anyway.
Current rules:
https://contributors.gettyimages.com/article/9608
and this
https://contributors.gettyimages.com/article/10847
-
I NEVER had any issues in spite of using topaz for denoising and sharpening routinely. But I uncheck the option to use their program-specific file prefix and my vote goes to simply not using their naming convention.
-
And what about Topaz Video Ai? On istock etc.
Is it possible to denoise movies, improve the quality of movie details like a Proteus filter without upscaling or improve slow motion?
-
Maybe strip the meta or exif data before submitting.
Removing the corresponding metadata (Exif tag ‘Software used’) could work.
Here is a screenshot of our own Zoom software where you can see that Topaz leaves an entry that is possibly recognised by IStock and is considered AI-generated.
In fact, this is technically correct, because the AI system is very similar to a Ki image generator. In the end, not a single original pixel of your photo is retained - the image is generated block by block in a different resolution/de-noised/business.
Strictly speaking, it is no longer your image, but a completely generated image to which you no longer have the rights.
If you want to avoid such problems, you would have to avoid scaling/de-noising/sharpening programmes with generative AI and rely on other algorithms.
Examples of such scaling programmes with ‘non-generative AIs’ would be Zoom #2 professional or Photozoom #8 and in the area of denoising, Neat Image or Denoise #5 professional come to mind.
Topaz Ai with all its Ai tools claims you own the copyright.
https://community.topazlabs.com/t/topaz-photo-commercial-usage/55873/4
Those of your recommended scaling programs also ashampoo zoom and zoom pro have Ai.
-
Those of your recommended scaling programs also ashampoo zoom and zoom pro have Ai.
That is correct, but the training data used in Zoom #2 Pro and Ashampoo Zoom #2 are only self made images.
How do I know that?
How do I best express this... I developed the Zoom #2 programme, so I can say with certainty that no data with third-party copyright was used.
-
So why wouldn't Topaz Ai also legally license the photos to Ai. This company is big and has a million users. A lot of people are making money from these Ai tools, but istock is still not clear about these Ai tools.
I will be happy to test these Ashampoo programs, they are also interesting.
-
So why wouldn't Topaz Ai also legally license the photos to Ai. This company is big and has a million users. A lot of people are making money from these Ai tools, but istock is still not clear about these Ai tools.
I will be happy to test these Ashampoo programs, they are also interesting.
The difference is in the way the AIs are used, but of course you can't see that from the outside.
Self-learned generative AI models require vast amounts of images to achieve good results with such a system, and by that I mean ranges of several hundred million images as a lower limit.
The AI system used in Zoom #2 pro is a non-generative system, which takes the information exclusively from the current image to be scaled (+ training images that can be specified additionally). This means that no large training volumes are required, but no new details are generated into the images.
Unfortunately, the topic is very complex and you need a lot of ‘basic knowledge’ if you want to make well-founded statements here.
I did not want to claim that the various providers have not all acquired the training images legally, but you should at least remain sceptical.
The large image datasets that can be ‘acquired’ from the relevant companies with 4-6 billion images cost 2-3 digit million sums - only the really big ones can afford such datasets...
Actually, I just wanted to say: ‘Remain sceptical’ - but I probably didn't manage that very well ;-)
-
Yes, I know it's worth being skeptical. Okay, but what will it change if, years later, it turns out that a well-known company that uses Ai tools for image scaling and enhancement was illegally using a photo database?
Would we authors also be complicit with stock agencies? After all, we would no longer remove improved photos or videos using Ai tools from our portfolio, which sell to customers, etc. By purchasing the Ai utility program, we should be released from liability. That's how I understand it. What is your opinion?
-
Yes, I know it's worth being skeptical. Okay, but what will it change if, years later, it turns out that a well-known company that uses Ai tools for image scaling and enhancement was illegally using a photo database?
Would we authors also be complicit with stock agencies? After all, we would no longer remove improved photos or videos using Ai tools from our portfolio, which sell to customers, etc. By purchasing the Ai utility program, we should be released from liability. That's how I understand it. What is your opinion?
It is currently very difficult or even impossible to answer this question. On the one hand, there are currently numerous lawsuits in court around the world concerning the use of images for AI training, some of which are being brought by artists, photographers, graphic designers, etc., because the big companies simply ‘download’ the entire Internet using web crawlers and use it as training data - but the problem has already been discussed extensively here.
Then there is the problem of the different regions, for example the EU has much stricter regulations than the USA, so does the company from the USA have to comply with the stricter rules for the EU customer of the software or not? Good question...
What happens now, if it turns out in the court cases (which I don't believe) that the companies are really ordered to use only copyright-free images in training and to prove this, would basically all images processed with AI tools (which were trained with unauthorised images/videos) not be legally flawless in the first place. In this case, the software customer would then have to claim the corresponding costs from the software provider (who used unauthorised data) - what a mess that would be, even if it were the right way :)
And to avoid such problems with my software, I don't use such tools, even if you can get some of them for free on the net and integrate them into commercial products.
Just because you can do it and it's possible doesn't mean you should do it :)
-
And to avoid such problems with my software, I don't use such tools, even if you can get some of them for free on the net and integrate them into commercial products.
Just because you can do it and it's possible doesn't mean you should do it :)
So you wouldn't even trust RAW file denoising programs because of their Ai tools? Like DxO PhotoLab or Pure Raw. They are the leaders and they boast that they have databases of billions of photos.
-
So you wouldn't even trust RAW file denoising programs because of their Ai tools? Like DxO PhotoLab or Pure Raw. They are the leaders and they boast that they have databases of billions of photos.
If you ask me privately, I generally only believe what I can verify - but as a mathematician, that's my own personal quirk ^^ Basically, I see every generative AI (trained with ‘billions of images’) as a potential legal problem in the future, but I'm also very cautious.
The results of these generative AI algorithms are impressive and certainly usable, in some cases (I would not denoise a night sky with stars with a generative AI, then I have some new stars in the sky afterwards and with luck also a few extra comets ;-)
At the moment, everything is unclear from a legal point of view (for example, it is not at all clear where ‘Sora’ got all the videos for their training) - committing to something here is currently on shaky ground.
-
Yes, but can photos after editing with DxO PhotoLab or Pure Raw be checked with any tools to see if it leaves any traces after using the Ai tool? Same with Topaz Ai.
Now Topaz Ai boasts that Gigapixel will be on iOS soon. Top brands such as NASA, Google, Tesla are also reportedly using these Ai tools from this company. Which somehow Topaz Ai are becoming more and more credible.
Yes, but Sora creates images from scratch from other images and there is a greater risk here. I'm more interested in the safety of photos and videos improved using Ai tools.
-
Yes, but can photos after editing with DxO PhotoLab or Pure Raw be checked with any tools to see if it leaves any traces after using the Ai tool? Same with Topaz Ai.
You can recognise from the image itself (with a certain degree of certainty) whether it was a generative AI that created the image or not. There are algorithms that search for typical artefacts in the image that result from the use of generative AI. This is therefore technically possible even without looking at Exif data or embedded watermarks - only the agency itself knows whether an agency actively uses this.
The trace is therefore the image itself or the specific arrangement of certain structures.
And yes, in order to recognise this via the pixels, an AI is used that is trained to distinguish between a ‘photo and an AI image’ - we have probably all experienced this before, where a real photo was rejected because it was supposedly taken with AI. In these cases, the ‘recognition AI’ was simply wrong :)
Yes, but Sora creates images from scratch from other images and there is a greater risk here. I'm more interested in the safety of photos and videos improved using Ai tools.
Yes, that's right and something very similar happens when you denoise with generative AI.
Let me try to describe the process (very simplified) for denoising:
-> the process takes a block of 32x32 pixels with image noise from the original and generates a block as similar as possible without image noise and adds it there, then it continues with the next block in the original image, etc.
The resulting image is therefore completely newly generated and may well contain deviations from the original (see example ‘additional stars in the sky’) - regardless of whether we denoise, sharpen or scale with these tools, the technology is always exactly the same, only the training data differs.
-
When it comes to artificial intelligence images that are created in generators from scratch, there are certainly algorithms that detect them.
https://sightengine.com/detect-ai-generated-images
And I doubt that our photos or videos corrected in Ai tools will be detected by any algorithms. Just add some noise to your photos or videos in a graphics or video program. No chance.
These brands, Topaz Ai or Dxo photolab, I doubt that they want to expose themselves to damage to the creators or any lawsuits. Creators have paid, so they use these Ai tools to make their creative work easier and better. What would be the point of getting into trouble with some free Ai tools.
-
Not so sure about that. Getty has already warned multiple times about using AI in their submissions and given plenty of advice of what can be done and what not. I would not be surprised if they close accounts that violate their policies , no matter how many files in their portfolios or how successful.....
But next generation cameras will be using ai integrated into their software.
What happens then?
AI will be everywhere. The agencies are fighting a losing battle. I'm sure iStock and others are already flooded with AI-modified videos and images without them even knowing it.
-
Graphic animators who buy, textures or matcap for their graphics and renders. How do they know that these derived textures or matcap are created in Ai generators? Or are they taking a risk as to Getty?
-
Resubmitted one of the files without the "Topaz AI" extension on the title and it has been accepted.
-
Getty replied to my ticket:
Topaz is defined as an AI tool. Please ensure any files edited using AI tools falls under our requirements.
Creative Content Retouching and Modification Requirements:
https://contributors.gettyimages.com/article/10847
I will go ahead and close this ticket. If you would like to reopen this ticket, please reply to the email notification of my response as replies via Account Management will not register a response if the ticket is closed. If you have another question or issue, please create a new ticket as we often need to assign tickets to different people and departments.
Sincerely,
Maggie
Find out more about how we license your work:
https://contributors.gettyimages.com/article/6251
Contributor Services
-
Getty replied to my ticket:
Topaz is defined as an AI tool. Please ensure any files edited using AI tools falls under our requirements.
Creative Content Retouching and Modification Requirements:
https://contributors.gettyimages.com/article/10847
I will go ahead and close this ticket. If you would like to reopen this ticket, please reply to the email notification of my response as replies via Account Management will not register a response if the ticket is closed. If you have another question or issue, please create a new ticket as we often need to assign tickets to different people and departments.
Sincerely,
Maggie
Find out more about how we license your work:
https://contributors.gettyimages.com/article/6251
Contributor Services
What did that actually say? Can you use Topaz or not, or only some parts of it?
"Did Not Connect: Potential Security Issue
Firefox detected a potential security threat and did not continue to contributors.gettyimages.com because this website requires a secure connection."
-
Getty replied to my ticket:
Topaz is defined as an AI tool. Please ensure any files edited using AI tools falls under our requirements.
Creative Content Retouching and Modification Requirements:
https://contributors.gettyimages.com/article/10847
I will go ahead and close this ticket. If you would like to reopen this ticket, please reply to the email notification of my response as replies via Account Management will not register a response if the ticket is closed. If you have another question or issue, please create a new ticket as we often need to assign tickets to different people and departments.
Sincerely,
Maggie
Find out more about how we license your work:
https://contributors.gettyimages.com/article/6251
Contributor Services
What did that actually say? Can you use Topaz or not, or only some parts of it?
"Did Not Connect: Potential Security Issue
Firefox detected a potential security threat and did not continue to contributors.gettyimages.com because this website requires a secure connection."
I asked why my images using Topaz solely for noise reduction and/or sharpening were rejected.
-
Also, I asked Istock if it is possible to enhance images with Topaz Ai or Dxo photolab tools.
I got a not clear answer. Does anyone understand this answer?
It is possible to use these Ai tools for image and video enhancement or not?
"
The majority of image editing software packages and smartphone platforms are incorporating generative AI technology. Each company may describe their tools in different ways, but they will often use words like ‘magic’, ‘generative’ ‘intelligence’, ‘assistant’, etc., when using generative AI. Also, if the technique you’re applying involves minimal effort or can be completed with a text prompt, the tool is likely to be using generative AI.
If in doubt, please assume that generative AI is being used and stick to the Retouching Requirements when using these tools. If you choose to follow our Modification Requirements, then it’s your responsibility to ensure that you are using traditional, non-generative AI tools.
AI tools may only be used if they comply with our retouching guidelines, however only whole image color adjustments are acceptable if you are retouching more than 10% of the images pixels.
https://contributors.gettyimages.com/article/10847
I will go ahead and close this ticket. If you would like to reopen this ticket, please reply to the email notification of my response as replies via Account Management will not register a response if the ticket is closed. If you have another question or issue, please create a new ticket as we often need to assign tickets to different people and departments."
-
Getty replied to my ticket:
Topaz is defined as an AI tool. Please ensure any files edited using AI tools falls under our requirements.
Creative Content Retouching and Modification Requirements:
https://contributors.gettyimages.com/article/10847
I will go ahead and close this ticket. If you would like to reopen this ticket, please reply to the email notification of my response as replies via Account Management will not register a response if the ticket is closed. If you have another question or issue, please create a new ticket as we often need to assign tickets to different people and departments.
Sincerely,
Maggie
Find out more about how we license your work:
https://contributors.gettyimages.com/article/6251
Contributor Services
What did that actually say? Can you use Topaz or not, or only some parts of it?
"Did Not Connect: Potential Security Issue
Firefox detected a potential security threat and did not continue to contributors.gettyimages.com because this website requires a secure connection."
I asked why my images using Topaz solely for noise reduction and/or sharpening were rejected.
I still don't understand the answer? :) Is Topaz allowed or not, or was one of the limitations in "retouching" the problem? But if you remove the name Topaz in the name of the file, then you're OK?
Do not add new elements, ‘inpaint’, or extend the canvas/‘outpaint’.
Do not retouch more than 10% of the image’s total pixels (whole image color adjustments are acceptable).
Is de-noising or sharpening, more than 10% of an image. I really don't see the answer in their advice page.
-
"When an image is created or edited using C2PA-compatible software, a set of cryptographically signed metadata is embedded directly into the image file. Content: This metadata includes information such as the creation time, the software used, and any edits made to the image."
In LR at least, Content Credentials are an opt-in menu option and not on by default. I do use Topaz AI for maybe 2 in 100 images if it needs extra work on the noise or sharpening front but havent yet had one rejected for it. My normal workflow is export from LR with metadata except keywords and description etc stripped (i do this by default to stop anyone scraping anything i dont add myself).
I did play with content credentials but couldnt really get it to work or do anything so i just disabled it again. No idea if PS is an opt-in or in by default.
-
I also received a response from Istock/Getty Images that they do not accept 3d graphics or animations that have derivative Ai textures.
''We do not accept AI generated content, this includes files created using AI generated elements, such as texture.''
This is a problem because when you buy textures to create your own 3d graphics and animations, you don't know if they are Ai textures.
-
You can't get pee out of a pool, and you can't get AI out of everyone's workflow. Agencies will need to figure out a way to embrace AI at some level eventually.
While I understand their issue with submitting 100% AI images or videos, I think you should be able to use AI as a tool to enhance camera-generated imagery.
As more mainstream productions are permitting AI content in projects, stock will need to adjust accordingly -- with appropriate labels so buyers can opt-out of browsing content that may have been "touched" by AI.
-
I wouldn't risk using Topaz AI in my workflow. It's very easy to get your account banned on stocks. It's not worth it.
-
But how will agencies detect that the photo or videos were de-noised and sharpened in Topaz Ai? If you enhance videos in Topaz Ai and then edit in Vegas Pro to Prores or another codec then there will be no Topaz Ai information. Topaz Ai does not take away the image copyright. It is not an image generator from scratch with prompt.
-
But how will agencies detect that the photo or videos were de-noised and sharpened in Topaz Ai? If you enhance videos in Topaz Ai and then edit in Vegas Pro to Prores or another codec then there will be no Topaz Ai information. Topaz Ai does not take away the image copyright. It is not an image generator from scratch with prompt.
If you export out of Topaz AI, Topaz will add tags to the file's metadata. If you edit the video and re-export and your exporter is set to maintain existing tags, the Topaz markings will remain.
Question is, are there ways other than these metadata "tags" for a system to determine what was used to edit the video throughout it's workflow?
-
Toppaz is advertising widely vauting that all their products are "powered by AI"
https://tinyurl.com/825jf28z (https://tinyurl.com/825jf28z)
iS especially can be uber-conservative with rejetions for IP, but also can be uber-inconsistent.
-
Yes Topaz Ai leaves tags however in Vegas Pro you can clear these tags after saving in ProRes codec without setting etc.
I doubt there is technology that detects if an image has been enhanced in Ai programs and how edited.
-
Yes Topaz Ai leaves tags however in Vegas Pro you can clear these tags after saving in ProRes codec without setting etc.
I doubt there is technology that detects if an image has been enhanced in Ai programs and how edited.
If you go down the path of violating stock requirements and trying to fool them, I think this is not the right path. Good luck to you!
-
I wrote to Topaz Ai asking if I have the copyright after image enhancement like simple dehazing or sharpening.
The company replied that I still have the copyright and that it is not a graphics generator with descriptions from scratch. I am not interested in scaling the image.
After all, the main requirement of stock agencies is to own the copyright, and I don't know if they should care how I edit my photos or videos. There is also such a thing as professional secrecy.
Istock/Getty Images previously didn't exactly answer my question about whether I can enhance images with Ai.
-
I would also add that on the Istock/Getty Images website they don't ban anything. Using Ai tools to remove noise, improve image quality like sharpening and even scaling. They also don't prohibit editing videos as far as improving image quality or adding smooth slow motion with Ai tools.
Istock/Getty Images mainly only writes about AI tools like cutting objects and pasting to about 10% and banning the upload of images of graphics, photos and videos created from scratch with generator descriptions.
"Which generative AI tools can I use to retouch my content? (Are the following tools allowed - Topaz denoise, Topaz upscale, Lightroom denoise, Lightroom lens blur, Photoshop Render a lens flare, etc?
Software packages and the specific tools/features within them will evolve over time, so we will not be providing guidance at this level. Instead, we’ve provided requirements for applying either Retouching or Modification. If you follow our Retouching Requirements, you do not need to worry about whether the tool is traditional or using generative AI. Alternatively, if you follow our Modification Requirements, it is your responsibility to be sure that you are not using generative AI tools. "
"Do NOT submit content created or modified using AI generative tools or models (e.g., Stable Diffusion, Dall-E, MidJourney, Adobe Firefly, etc.)"
https://contributors.gettyimages.com/article/10847
https://contributors.gettyimages.com/article/9608
-
Half the problem is the misuse/misunderstanding of the word "AI".
Its become a marketing slogan just as years ago everyone added "HD" after everything to make it sound cool.
AI in terms of sharpening/denoise etc is just a marketing term for "algorithm". We've always used algorithms in photo and video work to produce and edit images. Nothing has changed.
Thats different from *Generative AI* where the algorithm invents new things that weren't there at all instead of basing calculations on the content of the file itself.
Generative AI is frowned upon and can lead to accidental copyright infringement. Algorithm stuff basically cant.
Adobe have a fairly clear page on what constitutes AI use for flagging and what doesn't and its a fairly sensible, if hard to police, guideline.
-
Strangely, iStock is now having no issues accepting my Topaz sharpened/noise-reduced images.
-
However, ... if you paste an AI-generated image (Image A) over an unedited image (Image B) in Photoshop, flatten it, and save the result, Image B’s EXIF metadata will not retain or include any metadata from Image A. I think it will have NO data from image A, except the pixels copied. Basta. All the info, in the EXIF or out of it, will be the one Image B had from the beginning, except "edited in Photoshop" will be added almost for sure.
Yes, this works. I tried it, just for curiosity. It certainly wouldn't be worth the effort of doing this regularly.
It confirms that iStock is doing the most simple-minded thing possible. They just discriminate against the use of particular software tools because they are capable of doing generative AI, even if they are not being used for that purpose.