← Technology
Judge throws out Musk's xAI trade-secret suit against OpenAI: theft of secrets, or a baseless grudge?

Judge throws out Musk's xAI trade-secret suit against OpenAI: theft of secrets, or a baseless grudge?

A US federal judge dismissed — with prejudice — Elon Musk's xAI lawsuit accusing rival OpenAI of stealing trade secrets for chatbots. xAI alleged OpenAI poached its engineers and induced a former senior engineer to hand over confidential information about its Grok chatbot; Judge Rita Lin ruled xAI failed to back that up, calling further litigation 'futile.' It is Musk's second courtroom loss to OpenAI in four weeks, after a jury rejected his $150bn 'stealing a charity' claim in May.

The summary above is a neutral framing. Below, each side reports the same story in its own words — judge for yourself.

xAI / Musk: OpenAI stole our secrets

xAI's suit, first filed last September, accused OpenAI of misappropriating confidential information — alleging that engineers who left xAI for OpenAI took source code and trade secrets, and that OpenAI induced former senior engineer Xuechen Li to divulge confidential details about xAI's Grok chatbot. Musk's camp framed it as a rival unfairly building on stolen work; the amended complaint centered on a presentation Li gave while OpenAI was recruiting him.

OpenAI & the court: no case

US District Judge Rita Lin in San Francisco dismissed the lawsuit with prejudice — meaning it cannot be refiled — after already tossing an earlier version in February, saying it would be 'futile' to continue. She found xAI failed to show that OpenAI induced Li to divulge confidential information or that OpenAI's engineers knew he might have. The ruling is a vindication for OpenAI and Sam Altman, and Musk's second legal defeat against the company in a month.

More in Technology

US orders Anthropic to pull a frontier AI model over a 'jailbreak': vital security move, or disproportionate overreach?
Technology Jun 14

US orders Anthropic to pull a frontier AI model over a 'jailbreak': vital security move, or disproportionate overreach?

Anthropic disabled two newly launched AI models — Claude Fable 5 and the underlying Mythos 5 — to comply with a US export-control directive citing national security; its other models, including Claude Opus 4.8, are unaffected. New reporting sharpens the dispute: the White House's move was reportedly driven by suspicions that a China-linked group had gained access to Mythos, which officials fear could be reverse-engineered. Anthropic says the trigger it was told about was a narrow, minor 'jailbreak' replicable on other uncontrolled models, calls the recall disproportionate, and is fighting to reverse it.