OpenAI Erases Key Data in New York Times Copyright Case

The accidental deletion of evidence forces The New York Times to restart critical work in its lawsuit over AI training data.

The New York Times and other publishers are suing OpenAI and Microsoft, alleging their AI models were trained on copyrighted content without permission.
OpenAI engineers accidentally deleted data that the plaintiffs had spent over 150 hours compiling as potential evidence in the case.
Although OpenAI recovered some of the deleted data, the original file structure and names were lost, rendering the recovered data unreliable for use in court.
The New York Times' legal team stated they have no reason to believe the deletion was intentional but called the incident a significant setback in their efforts.
The plaintiffs are requesting that OpenAI conduct searches of its own datasets, arguing the company is better positioned to identify copyrighted material.