NYT wins now.

Timothy Lee and James Grimmelmann, “Why The New York Times might win its copyright lawsuit against OpenAI,” 02-20-2024, https://arstechnica.com/tech-policy/2024/02/why-the-new-york-times-might-win-its-copyright-lawsuit-against-openai/

If generative models never regurgitated copyrighted material, then defendants would have a compelling argument that it is transformative. The fact that the models occasionally produce near-perfect copies of other people’s creative work makes the case more complicated and could lead judges to view these companies more skeptically.

Conclusion

Generative AI developers have some strong arguments in response to copyright lawsuits. They can point to the value that their AI systems provide to users, to the creative ways that generative AI builds on and remixes existing works, and to their ongoing efforts to reduce memorization.

But all of these good arguments have something in common: they take copyright issues seriously. These responses acknowledge that generative AI is built on a foundation of training data, much of which is copyrighted, and then try to show that all of this copying is justified rather than that it is irrelevant.

The AI community is full of people who understand how models work and what they're capable of, and who are working to improve their systems so that the outputs aren't full of regurgitated inputs. Google won the Google Books case because it could explain both of these persuasively to judges. But the history of technology law is littered with the remains of companies that were less successful in getting judges to see things their way.