UPDATE All Deviations Are Opted Out of AI Datasets

3 min read

Deviation Actions

team's avatar
By
Published:
144.7K Views1 Collected Privately

UPDATE: We heard the community feedback, and now all deviations are automatically labeled as NOT authorized for use in AI datasets.

AI generators were trained using datasets collected from the open web. This includes content from creator platforms like DeviantArt, Pinterest, Twitter, and more. This was done without DeviantArt's permission and without your permission.


To help prevent your work from being used by third party AI datasets without your consent, we implemented a flag to tell AI datasets not to use your art.

All deviations on the platform are not authorized for inclusion in third-party datasets used to train artificial-intelligence models — unless you choose to opt in. DeviantArt has introduced the noai and noimageai directives to unambiguously communicate absence of such authorization. When an artist declares that their content may not be used as input to AI models, an HTML tag with the noimageai directive is placed on the page and an HTTP header with the noai directive is sent when the image is directly downloaded from DeviantArt’s servers. In order to remain in compliance with DeviantArt's updated Terms of Service, third parties that continue to use DeviantArt-sourced content to train machine-learning models of any kind must ensure their training data set excludes all content for which either of these directives are present. DeviantArt encourages other creator platforms to adopt this approach in order to ensure artists remain able to share their work with online audiences while retaining control over non-human usage.

IMPORTANT:

  • DeviantArt did not, does not, and will not use deviations submitted to DeviantArt to add to third-party AI models or training sets, or on DreamUp itself.

  • DeviantArt did not consent to third-party technology usage of images on our site, which were used to train AI models. In an effort to combat future unauthorized usage, we have enacted “noai” — an industry-first directive alerting AI models if deviants request for their work not to be used.

  • DeviantArt is the only platform giving creators the ability to tell third-party AI datasets and models whether or not their content can be used for training. This is a protection for creators to help them safeguard their content across the web.

  • We encourage other platforms who enable creators to publish their work to implement similar protections. We are making our “noai” directive available for use by all.

For Artists: How To Tell AI Datasets Not To Use Your Content

We're adding the ability to flag all your deviations, telling AI datasets they can't use your content. This is selected by default.

Screen Shot 2022-11-11 at 5.14.17 PM

For Developers: Implementing noai or noimageai

Developers can implement these directives in two ways.

Meta Tag

<meta name="robots" content="noimageai"> and/or <meta name="robots" content="noai">

HTTP Response Header

X-Robots-Tag: noimageai and/or X-Robots-Tag: noai

© 2022 - 2024 team
Comments3990
Join the community to add your comment. Already a deviant? Log In
Prolidon's avatar

Deviantart, are you trying to trick us? To trick us, your customers?


Here, you made a commitment to make to default that a person would not give consent to train an AI with their images unless they specified otherwise. This was right, because consent can only be considered true, when a person explicitly expresses it.

Screenshot 1
Screenshot 3

Now, with the new update, the default is that you express your consent to use the images for this purpose every time you post something, unless you choose otherwise. This will cause a lot of people to inadvertently post images that give consent for their material to be used for training, when they don't intend to give it.

Screenshot 2

Among other things, while previously the word regarding whether or not to give consent was clear, it is now expressed through the easily misunderstood writing "NoAI". What does NoAI mean? That this user is against artificial intelligence? That the image hasn’t been done with artificial intelligence? You can easily understand how misunderstandable it is and that only a portion of users know that "NoAI" means that the insertion of that image into the datasets to train the artificial intelligence is not allowed where the image of an unaware user could end up .


Am I supposed to believe that you did it all on purpose, hoping that many unsuspecting users would fall for it? I hope not, maybe it's just a mistake.


We want that consent is denied by default, unless otherwise expressed.


This is also for to the fact that you have made a commitment, as I quote here, a commitment to your customers.