Artificial Intelligence copyright litigation is still brewing in the courts. The questions about how generative AI models are trained are central into a wave of high-stakes lawsuits involving authors, publishers, and some of the world’s largest technology companies.
A historic $1.5 billion settlement by Anthropic delivered the first significant judicial ruling on the issue. The ruling is expected to influence how courts, litigants, and experts evaluate similar claims in the growing number of AI copyright cases working their way through the legal system.
Against this backdrop, the inaugural episode of CRA Sessions: AI in Practice, examines the issues at the center of these disputes.
This episode’s host and Competition Vice President Keler Marku speaks with Abhishek Nagaraj, a CRA academic affiliate and associate professor at UC Berkeley’s Haas School of Business, about AI training, copyright, and fair use.
Drawing on his research on AI memorization and copyright, Professor Nagaraj explains why training, memorization, reproduction, and market substitution are distinct concepts, and why empirical evidence is critical to evaluating AI-related copyright claims.
The discussion examines the evidence needed to assess claims of market harm in AI copyright litigation. Drawing on economic analysis, empirical testing, and technical understanding of AI systems, the conversation explores how lawyers can develop evidence-based arguments around fair use, substitution, and the impact of AI-generated outputs on copyrighted works.
Tune in to the conversation.
About CRA Sessions: AI in Practice
CRA Sessions: AI in Practice is a podcast series that examines the legal, economic, regulatory, and technical issues emerging from the rapid adoption of artificial intelligence.
Through conversations with CRA experts and leading academics, the series helps counsel, business leaders, and policymakers better understand how AI is reshaping markets, institutions, and strategy.
Disclaimer
The views expressed in this podcast are the views and opinions of the authors and do not reflect or represent the views of Charles River Associates or any of the organizations with which the authors may be affiliated. Detailed information about Charles River Associates, a trade name of CRA International, Inc., is available at www.crai.com.
