He says his warnings about Grok’s risks made him a target and led to dismissal, just days before SpaceX’s planned IPO.
A former engineer at Elon Musk’s xAI filed a lawsuit in a California state court, alleging that he was fired from SpaceX’s division for expressing concerns about the risks of artificial intelligence to humanity.
Devin Kim states in the lawsuit, filed on Tuesday, that his efforts to implement safety frameworks for developing the Grok chatbot made him a target for the company’s leadership.
The lawsuit comes ahead of SpaceX’s planned Friday IPO – the largest in history.
“Mr. Kim has repeatedly complained that the misalignment of xAI’s AI safety priorities, especially regarding Grok, nearly guarantees that the Company will engage in illegal acts, from inciting discrimination to the dissemination of weapons of mass destruction,”
– lawsuit
xAI and SpaceX did not respond to requests for comment regarding the lawsuit.
The Center for AI Safety, an organization focused on AI risks, said last week that it had appointed Kim as its president.
Elon Musk founded xAI in 2023 as a safer alternative to OpenAI, which he helped establish more than a decade ago. Last month, a court dismissed Musk’s lawsuit alleging that OpenAI had strayed from its original mission to benefit humanity.
According to the lawsuit, Kim was among the first hires at xAI in 2024 and, a few months after joining, received one of the key leadership roles. Kim claimed that Musk expected the implementation of appropriate safety mechanisms, but his supervisor, xAI co-founder Jimmy Ba, refused to follow these directives.
Kim also states that Ba abruptly fired him last year ahead of a presentation on AI safety before the company’s leadership.
The lawsuit accuses xAI and SpaceX of retaliation and wrongful termination under California law and seeks damages.
SpaceX and other Musk ventures, including Tesla, have previously faced criticism regarding employee safety and autonomous systems – topics that continue to surface in public discourse.
In 2023 there were at least 600 on-the-job injuries at SpaceX, including crush injuries, amputations, electrical injuries, and one fatal incident; data had not been previously released. The company did not comment at the time, but in court documents and outside court it defended its safety and said it provides substantial safety training.
Ultimately, this story underscores the growing pressure on tech giants to be accountable for AI safety and ethics and highlights the need for transparent risk-management processes in the industry.