Authors:

  • Jeremiah Smith
  • Ruth Okediji

Excerpt

November 2, 2023, Video: "On October 30, Harvard Law School hosted a discussion on “Intellectual Property and the Dawn of Generative AI,” featuring Justin Hughes '86, the Hon. William Matthew Byrne, Jr. Distinguished Professor of Law at Loyola Law School, and Pamela Samuelson, Richard M. Sherman Distinguished Professor of Law at Berkeley Law and co-director of the Berkeley Center for Law and Technology, in a conversation moderated by IP expert Ruth Okediji LL.M. ’91 S.J.D. ’96, Jeremiah Smith, Jr. Professor of Law at Harvard. The event was part of the Harvard Law School Rappaport Forum series, established in 2020 to promote and model rigorous, open, and respectful discussion of vital issues facing the world. In 2023, ChatGPT and other Large Language Models captured the attention of the world. But these are only the tip of the spear for generative artificial intelligence technologies. These technologies are poised to transform industry, education, and social relations. They raise a large number of challenges for law, perhaps none more so than for intellectual property. We are seeing significant lawsuits in the U.S. and UK over the massive amounts of copyrighted materials ingested for the purposes of training generative AI models. There are also important legal questions associated with the output of generative AI, including: When do their outputs infringe copyright and how should responsibility be allocated between the AI developer and the user whose prompt caused the infringing output? Should generative AI works be copyrightable and, if so, how do we decide who is the owner or author? Will generative AI change the contours of what we protect in copyright, such as a human artist’s “style?” Will the law’s decisions on these and related questions disrupt the livelihoods of workers in creative industries? These technologies also pose challenges for patent, trade secrecy, and right of publicity law. For example, what should be patent laws standards for “obviousness” and “enablement” — concepts the law has always measured by the human mind — going forward? Should AI developers have to disclose the data sets on which they trained their models, or is this data a protectable trade secret? This Rappaport Forum will explore these and other questions, including how IP policymakers in the U.S., the EU, and in other jurisdictions have been responding to the rise of another wave — this time, perhaps, a tsunami — of disruptive technology. "