Judge Dismisses xAI Trade Secrets Lawsuit Against OpenAI, Allows Refiling

A federal judge has tossed xAI’s trade-secrets case against OpenAI, ruling that Elon Musk’s startup didn’t plead enough facts tying OpenAI itself to any theft—while still leaving the door open for xAI to try again.

U.S. District Judge Rita F. Lin granted OpenAI’s motion to dismiss with leave to amend, finding that xAI’s allegations largely focused on former employees’ behavior rather than misconduct by OpenAI, the sole defendant.

xAI filed the lawsuit last September, accusing OpenAI of orchestrating a “coordinated, unfair, and unlawful campaign” to steal proprietary technology through targeted employee poaching.

The complaint alleged OpenAI “by hook or by crook” induced eight former xAI engineers to misappropriate the company’s source code, training methods, and data center deployment strategies, offering multi-million dollar packages to engineers who then stole files within hours of their communications with OpenAI recruiter Tifa Chen.’

“xAI does not allege any facts indicating that OpenAI induced xAI’s former employees to steal xAI’s trade secrets or that these former xAI employees used any stolen trade secrets once employed by OpenAI,” Judge Lin noted.

Lin set a deadline of March 17 for xAI to file a revised complaint and barred it from adding new claims or parties without court permission.

In the order, the judge wrote that to get past the initial stage and reach discovery, a plaintiff must allege facts that, if true, show the defendant committed the wrongful act.

In a tweet, OpenAI said that it “welcomed” the judge’s decision, claiming that the lawsuit was “baseless” and “another front in Mr. Musk’s ongoing campaign of harassment.”

Ishita Sharma, managing partner at Fathom Legal, told Decrypt the ruling “is reinforcing the high bar for claiming corporate trade-secret liability when employee movement is involved. Mere ‘poaching’ isn’t enough without concrete links tying the employer to misuse.”

She said a revised complaint, if filed, “will need to be much more detailed” and could be forced to narrow in on “specific conduct by OpenAI or narrower allegations about individual ex-employees rather than the company broadly.”

In fast-moving AI markets, Sharma added, courts want “very specific, fact-based allegations” and “simply hiring former employees isn’t enough” without evidence of inducement or use.

xAI alleged in its suit that early engineer Xuechen Li, one of its first 20 hires, uploaded “the entire xAI source code base” to a personal cloud account last July while communicating with OpenAI recruiter Chen on Signal, met her virtually minutes later, and received a multi-million-dollar offer by July 28, which he accepted by August 1.

Judge Rita F. Lin ruled that even if individuals engaged in misconduct, xAI failed to allege facts showing OpenAI directed, knew of, or used any trade secrets, noting Li never began employment after his offer was rescinded and that “mere possession of trade secrets is not sufficient,” and dismissed related state unfair competition claims as preempted.

A parallel case against Li remains active, with his response deadline extended to March 6 as both sides continue reviewing disputed files under a January preliminary injunction.

Musk vs OpenAI

The ruling is the latest setback in Musk’s legal campaign against his former venture, a relationship that has since become one of Silicon Valley’s most litigious rivalries.

Last year, xAI and X Corp. filed a federal antitrust lawsuit against Apple and OpenAI, alleging that an exclusive arrangement made ChatGPT the default AI on iPhones while locking out competitors like Grok.

Last month, Musk sought up to $134 billion from OpenAI and Microsoft, claiming entitlement to the “wrongful gains” they accrued from his early contributions.

Decrypt has reached out to OpenAI and xAI for comment.