News organizations, led by The New York Times, are seeking "serious sanctions" against OpenAI, accusing the AI firm of repeatedly misleading the court in a copyright infringement lawsuit. The core of the dispute centers on OpenAI's alleged concealment of evidence regarding users prompting ChatGPT to reproduce copyrighted articles, which could prove infringement or fair use.

During a re-deposition, OpenAI privacy engineer Vincent Monaco reportedly revealed the company lied for two years about the difficulty of searching ChatGPT logs. The AI firm allegedly claimed it lacked the technical ability to search large anonymized datasets, when in fact, it had already conducted such searches prior to litigation. The news plaintiffs argue this obstruction prolonged discovery, inflated expenses, and burdened the court.

OpenAI denies these allegations, suggesting the sanctions motion is a tactic by The New York Times to gain access to more user data as their case weakens. The news organizations counter that claims were streamlined, not dropped, and that OpenAI's deliberate withholding of 80 million log samples and subsequent over-redaction of a 20 million log sample demonstrates a wilful attempt to obstruct evidence. Sanctions sought include prohibiting OpenAI from using the heavily redacted log sample.