NYT Seeks Sanctions Over OpenAI Log Spat

OpenAI is in hot water after news organizations—led by The New York Times—accused the company of lying for years to hide evidence that its ChatGPT could regurgitate paywalled articles. In a sanctions motion filed Thursday, the plaintiffs claim OpenAI pretended it couldn’t search its logs to find evidence of users bypassing paywalls with chatbot prompts. But when an OpenAI engineer was re-deposed under court order, he inadvertently revealed the company had actually run such searches before the lawsuit even started. The news orgs argue this concealment wasted time and money, and they’re now calling for serious penalties. OpenAI fired back, calling the motion a desperate attempt to invade user privacy and noting that the Times recently dropped some claims from the case. This fight is critical because it goes straight to the heart of whether ChatGPT’s training on news articles is fair use or outright infringement—if OpenAI was hiding evidence, it could seriously damage its defense. Why it matters: This legal showdown could set a major precedent for how AI companies handle copyrighted content and whether they can be trusted to be transparent in court.