Former xAI engineer sues company over AI safety concerns, claims he was laid off for speaking up

A former engineer at Elon Musk's artificial intelligence venture xAI has taken legal action against the company and its parent firm SpaceX, alleging he was forced out after repeatedly warning that the company's chatbot, Grok, was being developed without adequate safeguards.

Devin Kim, who departed xAI in September 2025, filed a complaint in a California state court this week. The case arrives at a sensitive time, just days before SpaceX is expected to enter public markets in what is anticipated to become the biggest initial public offering in history.

Warnings over Grok's behaviour and harmful content

According to the lawsuit, Kim emerged as one of the strongest advocates for AI safety inside xAI while working on Grok. He is said to have repeatedly voiced concerns that the chatbot could fuel discrimination and potentially provide information linked to weapons of mass destruction.

The complaint argues that some of those fears later materialised. In one passage, the filing states: "Grok, of course, proved Mr Kim right by engaging in spectacular displays of online hatred and vitriol, with the model likening itself to Hitler ('MechaHitler')."

More from Tech

After that controversy, Kim allegedly helped reassess the chatbot's political leanings and discriminatory tendencies. Months after his departure, Grok once again attracted criticism when users employed the chatbot to spread non-consensual sexually explicit imagery across X, Musk's social media platform that is also part of the broader xAI ecosystem.

The lawsuit portrays Kim as a whistleblower who believed xAI's approach to AI safety created legal risks in areas ranging from consumer protection and internet regulations to rules governing arms and explosives.

Neither xAI nor SpaceX immediately commented on the allegations.

Finger pointed at former xAI co-founder Jimmy Ba

Interestingly, the lawsuit does not blame Musk for the alleged failures. Instead, Kim's legal team claims the billionaire had instructed the company to comply with laws and establish appropriate testing and safety procedures.

The complaint instead singles out former xAI co-founder Jimmy Ba, who exited the company earlier this year. Kim alleges that Ba dismissed his concerns, retaliated against him and attempted to suppress repeated complaints regarding safety and bias.

According to the filing, Ba was focused primarily on ensuring xAI won the race towards superintelligence. The lawsuit even claims that at one stage he remarked that "AI will kill us all anyway", suggesting safety concerns were secondary.

In another allegation, the complaint claims that during preparations for the release of Grok Code 1 in August 2025, Ba sought to avoid European Union safety requirements by allegedly misrepresenting aspects of the model to bypass mandatory testing.

The lawsuit states that Ba "would rather release an unsafe model than a poor-performing one", adding that Musk ultimately stepped in.

Kim had reportedly planned to present his findings during the week of September 15, 2025. However, before the presentation could take place, he was called into a meeting where, according to the complaint, Ba informed him they should "go [their] separate ways" without offering a convincing explanation.

Kim's interest in AI safety predates his time at xAI. During his stint at Scale AI, he worked on projects aimed at creating training data to help AI systems identify harmful content and comply with governance standards. Last week, the non-profit Centre for AI Safety appointed him as its president.

Through the lawsuit, Kim is seeking compensatory and punitive damages, along with a declaration from the court that the actions of xAI and SpaceX were unlawful.

The case is likely to intensify scrutiny over how leading AI firms balance the pressure to innovate quickly with the need to ensure their systems remain safe and compliant with regulations.