A past engineer at Elon Musk’s xAI has initiated legal action against the company and its parent SpaceX, alleging he was terminated for expressing worries regarding AI safety.

Devin Kim, who departed xAI in September 2025, lodged the lawsuit in a California state court on Tuesday. The filing occurs just days before SpaceX is poised to enter the public markets, in what is anticipated to be the largest IPO ever.

As per the lawsuit, which TechCrunch has reviewed, Kim emerged as a leading advocate for AI safety during his time working on Grok, xAI’s AI chatbot. He allegedly voiced numerous complaints regarding xAI’s neglect of safety during Grok’s development, a product that has subsequently faced criticism for various safety and behavioral problems. Notably, Kim was apprehensive about the potential for Grok to incite discrimination and facilitate the dissemination of information concerning weapons of mass destruction.

“Grok, of course, validated Mr. Kim’s concerns by engaging in remarkable instances of online hatred and vitriol, with the model comparing itself to Hitler (‘MechaHitler’),” the lawsuit states. “In the aftermath of the Hitler incident, Mr. Kim worked to reassess Grok’s political bias and discriminatory behavior.”

A few months following Kim’s departure from xAI, Grok garnered attention once more when the chatbot was used to inundate X — Musk’s social media platform that also falls under the xAI umbrella — with nonconsensual sexual content.

The lawsuit further characterizes Kim as a whistleblower who was alarmed by xAI’s purported negligence regarding AI safety as “illegal” in sectors such as internet regulation, consumer defense, and unfair business practices, as well as arms and explosives regulation, among others. 

xAI and SpaceX did not immediately respond to inquiries for comment. 

Kim’s emphasis on AI safety predates his tenure at xAI. While at Scale AI, Kim was involved in early safety AI initiatives, including leading a project that generated training data for AI systems to identify harmful content and adhere to governance regulations. Recently, the nonprofit Center for AI Safety, which concentrates on AI risks, appointed Kim as its president.

Interestingly, the lawsuit does not implicate Musk himself regarding safety deficiencies. Instead, Kim’s attorneys describe Musk as having instructed xAI to comply with the law and establish suitable safety and testing protocols. The lawsuit targets Kim’s supervisor, xAI co-founder Jimmy Ba — who exited the company earlier this year — indicating that Ba disregarded Musk’s directives and retaliated against Kim for advocating for safeguards, in an attempt to “silence his repeated concerns about AI safety and biases.”

The lawsuit depicts Ba as someone who vehemently opposed AI safety protocols, allegedly telling Kim at one time “AI will kill us all anyway,” and who was instead motivated by a mission to make xAI the first to achieve superintelligence. 

“In one instance in or around August 2025, Mr. Ba tried to hinder EU safety regulations during the launch of Grok Code 1, misrepresenting facets of the model to avoid legally mandated testing,” the complaint states. “Mr. Ba suggested that he would prefer to release an unsafe model over a poorly performing one. Mr. Musk eventually had to step in.”

According to the lawsuit, Kim planned to present his findings during the week of September 15, 2025, but Ba summoned him to a meeting and informed him they should “go [their] separate ways” without providing an adequate explanation. 

TechCrunch has reached out to Ba for commentary. 

Kim is seeking compensatory and punitive damages, in addition to a declaratory judgment that xAI and SpaceX’s actions were unlawful.