Over the past two months, the Copyright Office hosted four public, virtual listening sessions on the use of artificial intelligence to generate creative works. The listening sessions focused on literary works, including print journalism and software; visual arts; audiovisual works, including video games; and music and sound recordings. Artists, creators, AI developers, researchers, lawyers, academics, and more shared their goals, concerns, and experiences related to the use and impact of generative AI.
As we look back on four vibrant sessions, we wanted to share some highlights for those who were not able to join live.
Register of Copyrights Shira Perlmutter opened the first listening session, on literary works, with a set of guiding questions for all four sessions: “How does current law apply? Should it be changed? [H]ow will the copyright community, from creators to users, be impacted?” She also reminded the audience that the Copyright Office plays a role “both in addressing practical concerns and in advising on policy.”
Following Register Perlmutter’s introductory remarks, participants spoke on two consecutive panels in which they articulated a wide-ranging set of perspectives.
The remaining listening sessions followed a similar format, and some included additional opportunities for comments without further discussion. Copyright Office staff moderated the listening sessions, and Associate Register of Copyrights and Director of Policy and International Affairs Maria Strong and General Counsel and Associate Register of Copyrights Suzy Wilson each made remarks.
At the final listening session, Register Perlmutter observed some of the themes of the series, including that
- there is disagreement about whether, or under what circumstances, training generative AI on copyrighted works could be considered fair use;
- there is considerable interest in developing methods to enhance transparency and education regarding how generative AI produces works, including the possibility of tracking relationships between ingested works and outputs, and understanding how assistive AI is used as a tool in the creation process; and
- many stakeholders still have questions about the Office’s registration guidance for works containing AI-generated material and would like more details and more examples of how the Office will approach applications for such works.
Throughout all four listening sessions, the Office heard from a broad and diverse group of stakeholders, experts, and creatives, including some who do not typically participate in Office roundtables. Among the speakers were a professor of computer and information science; several Academy Award-nominated artists; attorneys for major private actors, including tech companies and music streaming platforms; representatives from various unions, guilds, and trade groups; and independent visual artists, filmmakers, and composers.
The listening sessions broke registration and attendance records for Copyright Office events. Over 4,100 people tuned in over the course of the four sessions. In her final remarks, Register Perlmutter thanked all the panelists for sharing their insights and the public for tuning in. “The Office appreciates the high level of public engagement with these listening sessions,” said Register Perlmutter. “This interest is of course a reflection of the astonishing potential of artificial intelligence, and the impact that its already having in our lives and on society as a whole.”
The feedback and comments provided to the Office during the listening sessions will help guide the next steps in the Office’s AI initiative. The Office is drafting a notice of inquiry, to be published in the Federal Register later this summer, which will solicit written comments from the public on a wide range of issues involving AI and copyright. Issues raised during the sessions will directly inform the questions asked in the notice.
If you missed the listening sessions initially, or want to listen back again, you can review all the materials from each of the sessions on our website, including the agenda, transcript, and full video recording.