US Copyright Office Initiates Inquiry Into Regulation of Generative AI

The United States Copyright Office has taken on a critical mission to tackle copyright challenges posed by generative AI. This has initiated a query seeking public feedback to address copyright issues involving AI.

The move highlights the emerging importance of confronting copyright issues arising from the use of generative AI for creative purposes.

Last March, an AI initiative was launched by the Copyright Office to examine copyright infringement issues caused by generative AI technology.

The agency has already organized four public sessions and two webinars discussing the use of copyrighted creative material for generative tools.

The Register of Copyrights and Director of the US Copyright Office, Shira Perlmutter, explained that they had launched the initiative at the beginning of the year to sort out complex issues related to generative AI.

Besides, the agency has gathered some feedback and is not seeking any more input from “the broadest audience to date in the initiative”. The Copyright Office would be advising Congress to use this information and control its own regulatory work.

It would also provide courts, the public, and government entities with information and resources.

The Ownership Dilemma Over Copyrighted Materials Continues

With the extensive popularity of generative AI, one of the key issues of the technology lies in determining the ownership of AI-generated content.

A partner in the technology practice of Everest Group, Nitish Mittal, stated that technology providers often act as platforms for AI models but avoid taking a stance on content ownership.

This ambiguity has led to disputes, with figures like Sarah Silverman, Richard Kadrey, and Christopher Golden and the Writers Guild of America raising copyright infringement claims against generative AI.

Mittal also identified four primary risks associated with generative AI that need immediate attention. Given that AI systems rely on vast sets of data, it’s imperative to ensure the privacy and security of these data.

Besides, he emphasized that decision-making processes must be transparent and explainable so that they maintain accountability. Also, a complex challenge lies in determining the owner of AI-generated content.

He stated that fairness and ethical considerations should be prioritized while developing and using AI systems to prevent bias.

Copyright Office Actively Seeks Input From Various Stakeholders

The Copyright Office is actively seeking input from various stakeholders, including content producers, legal entities, and technology providers. The gathered feedback would be effective in establishing a common regulatory framework.

This ensures that the norms would be consistently incorporated.

Generative AI largely depends on training models with vast amounts of data, which includes copyrighted material.

Earlier this year, the Biden administration addressed the risks associated with generative AI. Vice President Kamala Harris and other officials also met with the CEOs of major tech companies, including Google, OpenAI, and Microsoft, to discuss guidelines without getting into any legal commitment.

Avivah Litan, a vice president analyst for Gartner, focused on how old-world copyright regulations and new-world innovations come to a clash.

She also suggested the scope of creating standards to identify copyright materials used in training AI models on the lines of the Coalition for Content Provenance and Authenticity (C2PA) for media content.

Litan further identified the financial requirements to implement these solutions and emphasized the importance of regulators setting policies and timeframes for the implementation.