Generative AI: The Unintended Consequences and the Rise of "Training-Defeating" Tools
It’s a wonder what generative AI, particularly text-to-image AI models like Midjourney and OpenAI’s DALL-E 3, can do. From photorealism to cubism, image-generating models can translate practically any description, short or detailed, into art that might well have emerged from an artist’s easel.
However, the trouble is that many of these models — if not most — were trained on artwork without artists’ knowledge or permission. While some vendors have begun compensating artists or offering ways to ‘opt out’ of model training, many haven’t.
The Need for "Training-Defeating" Tools
In lieu of guidance from the courts and Congress, entrepreneurs and activists are releasing tools designed to enable artists to modify their artwork so that it can’t be used in training GenAI models. One such tool, Nightshade — released this week — makes subtle changes to the pixels of an image to trick models into thinking the image depicts something different from what it actually does.
Another tool, Kin.art, uses image segmentation (i.e., concealing parts of artwork) and tag randomization (swapping an art piece’s image metatags) to interfere with the model training process. Launched today, Kin.art’s tool was co-developed by Flor Ronsmans De Vry, who co-founded Kin.art, an art commissions management platform, alongside Mai Akiyoshi and Ben Yu a few months ago.
How "Training-Defeating" Tools Work
As Ronsmans De Vry explained in an interview, art-generating models are trained on datasets of labeled images to learn the associations between written concepts and images, like how the word ‘bird’ can refer to not only bluebirds but also parakeets and bald eagles (in addition to more abstract notions). By ‘disrupting’ either the image or the labels associated with a given piece of art, it becomes that much harder for vendors to use the artwork in model training, he says.
The Importance of Artist Rights
An artist profile on Kin.art. Image Credits: Kin.art
‘Designing a landscape where traditional art and generative art can coexist has become one of the major challenges the art industry faces,’ Ronsmans De Vry told TechCrunch via email. ‘We believe this starts from an ethical approach to AI training, where the rights of artists are respected.’
Ronsmans De Vry asserts that Kin.art’s training-defeating tool is superior in some ways to existing solutions because it doesn’t require cryptographically modifying images, which can be expensive. But, he adds, it can also be combined with those methods as additional protection.
Kin.art’s model-defeating segmentation method. Image Credits: Kin.art
The Limitations of Existing Solutions
‘Other tools out there to help protect against AI training try to mitigate the damage after your artwork has already been included in the dataset by poisoning,’ Ronsmans De Vry said. ‘We prevent your artwork from being used in model training in the first place.’
The Rise of Kin.art and Nightshade
Launched this week, Kin.art’s tool is part of a growing trend of "training-defeating" tools designed to protect artists’ rights. The tool uses image segmentation and tag randomization to make it difficult for vendors to use artwork in model training.
Nightshade, released earlier this year, makes subtle changes to the pixels of an image to trick models into thinking the image depicts something different from what it actually does. While both tools have their limitations, they represent a growing effort by artists and entrepreneurs to reclaim control over their work in the age of generative AI.
The Future of Generative AI and Artist Rights
As generative AI continues to advance, the need for "training-defeating" tools will only grow more pressing. With vendors increasingly relying on artwork to train their models, it’s imperative that artists have a say in how their work is used.
While Kin.art and Nightshade represent a step in the right direction, much more needs to be done to protect artist rights in the age of generative AI. By developing innovative tools like "training-defeating" software, entrepreneurs and activists can help reclaim control over artistic creations and ensure that they benefit from their own work.
The Need for Regulation
However, regulation will also play a critical role in protecting artist rights. As AI-powered art becomes increasingly prevalent, lawmakers must consider the unintended consequences of generative models on the art world. By establishing clear guidelines around artist consent and compensation, policymakers can help prevent the exploitation of artists by vendors seeking to capitalize on their work.
Conclusion
The rise of "training-defeating" tools like Kin.art and Nightshade marks a turning point in the struggle for artist rights in the age of generative AI. As the art world grapples with the unintended consequences of these models, it’s imperative that we prioritize the protection of artistic creations and the artists who create them.
By supporting innovative solutions like "training-defeating" software and advocating for regulation that prioritizes artist consent and compensation, we can help ensure that the benefits of generative AI are shared equitably among all stakeholders. Only then can we unlock the full potential of this technology and celebrate its artistic achievements without compromising the rights of those who create them.
Recommended Reading
- "The Ethics of AI-Generated Art" by Dr. Joanna Bryson, University of Bath
- "The Unintended Consequences of Generative AI on Artist Rights" by Rachel C. Lee, Harvard Law School
- "AI-Powered Art: The Future of Creativity and the Need for Regulation" by David G. Kirp, Berkeley Law
Subscribe to Our Newsletter
Stay up-to-date with the latest news and insights on generative AI and artist rights. Subscribe to our newsletter today!