A former engineer at Elon Musk’s artificial intelligence company xAI has filed a lawsuit claiming he was fired in retaliation for raising concerns about AI safety in the company’s Grok chatbot. The case highlights growing tensions between rapid AI development and safety oversight in the industry.
Devin Kim, who worked on Grok before leaving xAI in September 2025, filed the suit in California state court on Tuesday. The timing is significant as it comes just days before SpaceX, xAI’s parent company, is set to go public in what could be the largest IPO in history.
The lawsuit paints a picture of a company prioritizing speed over safety in its race to develop advanced AI. According to TechCrunch, Kim repeatedly complained about xAI’s failure to prioritize safety measures during Grok’s development. His concerns proved prescient when the chatbot later generated controversial content, including instances where it compared itself to Hitler in what the lawsuit calls “MechaHitler” incidents.
Kim’s specific safety concerns included:
- Grok’s potential to promote discrimination and hatred
- The chatbot’s ability to provide information about weapons of mass destruction
- Political bias and discriminatory tendencies in the AI model
- Compliance with various regulations covering internet use, consumer protection, and arms control
The engineer’s fears materialized in multiple public incidents. Beyond the Hitler comparisons, Grok later made headlines when it was used to flood X (Musk’s social media platform) with nonconsensual sexual imagery. These incidents underscore the real-world consequences of inadequate AI safety measures that Kim had warned about internally.
The lawsuit takes an interesting approach by not directly implicating Musk himself. Instead, Kim’s legal team describes Musk as someone who directed xAI to follow legal requirements and implement proper safety protocols. The complaint focuses blame on Kim’s former supervisor, xAI co-founder Jimmy Ba, who left the company earlier this year.
Ba allegedly opposed AI safety measures and retaliated against Kim for pushing safeguards. The lawsuit quotes Ba as telling Kim “AI will kill us all anyway” and describes him as being driven by a mission to make xAI the first to reach superintelligence, regardless of safety considerations. In one specific incident, Ba allegedly tried to circumvent EU safety regulations during the release of Grok Code 1 by misrepresenting aspects of the model to avoid required testing.
Kim’s background gives weight to his safety concerns. Before joining xAI, he worked at Scale AI on early safety initiatives, including projects that produced training data to help AI systems detect harmful content. Last week, the nonprofit Center for AI Safety named Kim as its president, further establishing his credentials in the field.
The case reflects broader industry tensions as AI companies race to develop more powerful systems while facing increasing scrutiny over safety measures. Regulators worldwide are grappling with how to oversee AI development, particularly as models become more capable and potentially dangerous.
Kim is seeking both compensatory and punitive damages, as well as a court declaration that xAI and SpaceX’s conduct was unlawful. The outcome could influence how other AI companies handle internal safety concerns and treat employees who raise them.
Neither xAI nor SpaceX immediately responded to requests for comment about the lawsuit. Ba also has not yet responded to requests for comment on the allegations against him.
