U.S. Senator Peter Welch from Vermont recently introduced a new bill called the "Artificial Intelligence Network Transparency and Accountability Act" (TRAIN Act), aimed at increasing transparency for AI developers.
If passed, this legislation would allow copyright holders to issue subpoenas to relevant parties to obtain training records when they believe their works have been used without authorization to train AI models.
Image Source Note: Image generated by AI, image licensed by Midjourney
According to this bill, copyright holders only need to assert a "good faith belief" to request AI developers to provide sufficient training materials to determine whether their works were used. If developers fail to comply with this request, it will be legally presumed that they did use the relevant copyrighted works until they can prove otherwise.
Welch stated that as AI becomes increasingly integrated into the lives of Americans, it is essential to "set higher standards for transparency." He emphasized, "If your work is used to train AI, as a copyright holder, you should have a way to determine if it has been used and receive appropriate compensation if it has." He pointed out that artists, musicians, and creators need a tool to find out whether AI companies have used their works without permission.
In recent years, the rapid development of generative AI technology has raised numerous legal and ethical issues, particularly as artists worry that these tools may allow others to replicate their works without permission, attribution, or compensation. For example, a viral spreadsheet from the well-known AI art generation tool Midjourney listed thousands of artists and their works, further confirming artists' concerns.
Meanwhile, many companies that rely on human creativity have begun to take action against AI developers. Recently, news organizations such as The New York Times and The Wall Street Journal have filed infringement lawsuits against companies like OpenAI and Perplexity AI. The world's largest record companies also jointly sued two prominent AI music generation companies in June, claiming they used decades of copyrighted music records for model training without authorization.
As legal disputes escalate, over 36,000 creative professionals, including Oscar-winning actress Julianne Moore, author James Patterson, and Radiohead's Thom Yorke, have signed an open letter calling for a ban on the unauthorized use of human artistic works to train AI. Although the U.S. has yet to enact comprehensive federal laws to regulate AI development, some states have begun to push for AI-related legislation.
Welch noted that this bill has received support from various organizations, including the Screen Actors Guild, the American Federation of Musicians, and the Recording Academy, as well as some major companies. Nevertheless, with a busy congressional schedule and numerous priorities to address, the likelihood of this bill passing in the current Congress is low. Welch's office indicated plans to reintroduce the bill next year.
Key Points:
📄 Senator Peter Welch introduces a new bill aimed at enhancing transparency for AI developers and protecting the rights of copyright holders.
🎨 The bill allows copyright holders to request training records of AI models under a "good faith belief."
⚖️ The bill has garnered support from various arts and music organizations, despite challenges posed by a busy congressional schedule.
