OpenAI Again Defeats X.AI Trade Secrets Suit Over Code, Ex-Staff

OpenAI Inc. convinced a federal judge for the second time to dismiss X.AI Corp.'s lawsuit accusing the AI giant of poaching employees and encouraging them to steal confidential information, including source code.

Elon Musk’s X.AI failed to sufficiently allege trade secret misappropriation because asking a new hire to talk about past work doesn’t equate to encouraging theft of trade secrets, Judge Rita F. Lin of the US District Court for the Northern District of California said Monday. Lin’s current order doesn’t allow X.AI to amend its complaint again.

Lin initially dismissed X.AI’s suit in February, saying the company failed to demonstrate any misconduct by OpenAI and instead merely pointed to eight former X.AI employees who left for OpenAI at around the same time.

X.AI filed an amended complaint in March after an unsuccessful bid for a six-month extension to bolster its claims, triggering OpenAI to file another motion to dismiss. But Lin said the new allegations attempted to paint “routine” parts of the hiring process—such as asking about past projects and assessing where former X.AI engineer Xuechen Li would best fit—as nefarious efforts to induce trade secrets theft.

“To hold otherwise would potentially expose employers to liability any time they inquire about a candidate’s past work,” Lin wrote.

The legal fight began last August when X.AI sued its former AI researcher Xuechen Li, alleging that he admitted stealing trade secrets regarding its AI platform Grok to bring with him to OpenAI. Musk’s company escalated the fight the next month by filing a separate trade secrets lawsuit against OpenAI itself.

Munger, Tolles & Olson LLP represents OpenAI. Winston & Strawn LLP, King & Spalding LLP, and Wilkinson Stekloff LLP represent X.AI.

The case is X.AI Corp. v. OpenAI Inc., N.D. Cal., No. 3:25-cv-08133, 6/15/26.