A former engineer at Elon Musk’s xAI has filed go well with in opposition to the corporate and its guardian SpaceX claiming he was fired for elevating issues about AI security.
Devin Kim, who left xAI in September 2025, filed the go well with in a California state court docket on Tuesday. The grievance comes days earlier than SpaceX is about to join the public markets in what’s shaping as much as be the biggest IPO in historical past.
In accordance with the lawsuitwhich TechCrunch has seen, Kim grew to become a outstanding voice for AI security whereas engaged on Grok, xAI’s AI chatbot. He allegedly complained repeatedly about xAI’s failure to prioritize security in Grok’s growth, a product that has since come underneath hearth for a variety of security and behavioral points. Specifically, Kim was involved with the chance that Grok may foment discrimination and assist unfold details about weapons of mass destruction.
“Grok, in fact, proved Mr. Kim proper by partaking in spectacular shows of on-line hatred and vitriol, with the mannequin likening itself to Hitler (‘MechaHitler’),” the lawsuit reads. “Following the Hitler debacle, Mr. Kim labored to re-evaluate Grok’s political bias and discriminatory tendencies.”
A couple of months after Kim departed xAI, Grok made headlines once more when the chatbot was used to flood X — Musk’s social media platform that additionally falls underneath the xAI umbrella — with nonconsensual sexual imagery.
The lawsuit additionally positions Kim as a whistleblower who was involved about xAI’s alleged disregard for AI security as “illegal” in areas equivalent to web regulation, client safety and unfair enterprise practices, and arms and explosives regulation, amongst others.
xAI and SpaceX didn’t instantly reply to requests for remark.
Kim’s concentrate on AI security predates his time at xAI. Whereas working at Scale AI, Kim labored on early security AI initiatives, like main a mission that produced coaching information for AI to coach programs to detect dangerous content material and adjust to governance insurance policies. Final week, the nonprofit Middle for AI Security, which focuses on AI dangers, named Kim as its president.
Apparently, the lawsuit doesn’t implicate Musk himself as a cause for an absence of security. Quite, Kim’s attorneys describe Musk as having directed xAI to observe the legislation and implement applicable security and testing processes. As an alternative the declare targets Kim’s supervisor, xAI co-founder Jimmy Ba — who left the company earlier this year — saying that Ba ignored Musk’s directives and retaliated in opposition to Kim for pushing for safeguards, in an effort to “silence his repeated complaints about AI security and biases.”
The lawsuit portrays Ba as somebody who vehemently opposed AI security measures, allegedly telling Kim at one level “AI will kill us all anyway,” and who was as a substitute pushed by a mission to make xAI the primary to succeed in superintelligence.
“In a single occasion in or round August 2025, Mr. Ba tried to thwart EU security laws through the launch of Grok Code 1, misrepresenting points of the mannequin to be able to keep away from legally required testing,” the grievance says. “Mr. Ba indicated that he would somewhat launch an unsafe mannequin than a poor-performing one. Mr. Musk in the end needed to intervene.”
In accordance with the lawsuit, Kim meant to present a presentation of his findings the week of September 15, 2025, however Ba known as him into a gathering and advised him they need to “go (their) separate methods” with out offering a passable cause.
TechCrunch has reached out to Ba for remark.
Kim is in search of compensatory and punitive damages, in addition to a declaratory judgment that xAI and SpaceX’s conduct was illegal.
Whenever you buy by means of hyperlinks in our articles, we may earn a small commission. This doesn’t have an effect on our editorial independence.
