California District Judge Araceli Martínez-Olguín recently ruled in favor of OpenAI, dismissing the majority of claims made by writers Sarah Silverman, Michael Chabon, and Paul Tremblay. The trio alleged that the large language model behind ChatGPT had been trained on pirated copies of their books without consent. Here’s a breakdown of the key points from the ruling.
Lack of Substantiated Evidence:
Judge Martínez-Olguín emphasized that the writers failed to provide sufficient evidence supporting their claims. Allegations of indirect copyright infringement were deemed unsubstantiated, as the plaintiffs could not demonstrate “substantial” or “exact” similarity between ChatGPT responses and their original works.
Unproven Material Damage:
Despite the writers asserting that OpenAI’s actions caused them material damage, the judge highlighted the lack of concrete evidence. Even if the law was violated, the court found it challenging to determine the extent of the alleged damages.
Dismissal of Unjust Enrichment Claims:
The judge dismissed claims of unjust enrichment, citing the writers’ failure to specify how OpenAI benefited from training ChatGPT on their works. Allegations of negligence were also discarded.
In response, OpenAI plans to pursue claims of direct infringement in later stages of the proceedings, notes NIX Solutions. The writers are given until March 13 to amend their arguments, requiring evidence of ChatGPT responses resembling their work and proof that OpenAI intentionally facilitated copyright infringement by removing output data.