Amid a growing debate about copyright and ethics in generative artificial intelligence, a new non-profit, Fairly Trained, founded by former Stability AI executive Ed Newton-Rex, is introducing an initiative to evaluate and certify AI models based on their respect for creators’ rights.
The initiative, launched on Wednesday (January 17), aims to provide consumers and businesses with a transparent way to identify AI models trained on data acquired with consent from creators, potentially influencing the industry’s approach to copyright and creative rights.
Similar to organic food certifications, Fairly Trained’s “Licensed Model” certification evaluates and certifies AI models based on their training data sources.
This contrasts with the prevalent “fair use” argument employed by some AI companies, which allows for data scraping without explicit permission.
“In this way, we hope it will be a helpful marker of companies who are taking a fairer approach to training data acquisition.”
“In this way, we hope it will be a helpful marker of companies who are taking a fairer approach to training data acquisition,” Fairly Trained said.
Fairly Trained’s founder, Ed Newton-Rex, resigned from Stability AI in November 2023 due to concerns about their stance on fair use, believing it undermines creators’ rights in light of AI’s potential to commercially compete with copyrighted works.
The California-based entrepreneur led the development of Stable Audio, a generative AI music-making platform trained on licensed music in partnership with rights-holders. Prior to joining Stability AI, he founded music-making AI platform Jukedeck over a decade ago then sold it to TikTok in 2019. He then became Product Director of TikTok’s in-house AI Lab, before becoming Chief Product Officer at music app Voisey, which was sold to Snap in late 2020.
Upon leaving Stability AI, Newton-Rex said he “wasn’t able to change the prevailing opinion on fair use” at the company.
“Companies worth billions of dollars are, without permission, training generative AI models on creators’ works, which are then being used to create new content that in many cases can compete with the original works. I don’t see how this can be acceptable in a society that has set up the economics of the creative arts such that creators rely on copyright,” he said at the time.
“It’s clear that there are lots of people out there who would rather use gen AI models trained on licensed data, since this is a fairer way to treat the creators behind the data than the alternative – scraping and claiming it’s fair use,” Newton-Rex said in a LinkedIn post on January 17. “Our hope is that this certification can help people identify the most creator-friendly models to use.”
“This initiative counters the industry’s opacity in data procurement, ensuring companies meet rigorous ethical standards.”
air Adato, BRIA AI
Most recently, Fairly Trained said nine startups, primarily focused on sound generation with one image generation company, have already achieved Fairly Trained certification. These startups span image, music and singing voice generation firms including Beatoven.AI, Boomy, BRIA AI, Endel, LifeScore, Rightsify, Somms.ai, Soundful, and Tuney.
“We proudly support the Fairly Trained certification. This initiative counters the industry’s opacity in data procurement, ensuring companies meet rigorous ethical standards,” said Yair Adato, CEO at image generation company BRIA AI.
While still in its infancy, Fairly Trained has garnered support from organizations like the Association of American Publishers, the Association of Independent Music Publishers, Concord, Pro Sound Effects, and Universal Music Group.
“We welcome the launch of the Fairly Trained certification to help companies and creators identify responsible generative AI tools that were trained on lawfully and ethically obtained materials.”
Christopher Horton, Universal Music Group
“Universal Music Group has taken an industry-leading position in calling for responsible and ethical uses of AI technology that further enhance creativity, support artists and respect copyright. We welcome the launch of the Fairly Trained certification to help companies and creators identify responsible generative AI tools that were trained on lawfully and ethically obtained materials,” said Christopher Horton, SVP, Strategic Technology at Universal Music Group.
Music Business Worldwide