Two former OpenAI employees and a group of nonprofits focused on artificial intelligence security are warning that Elon Musk’s artificial intelligence lab, xAI, could become a liability for potential investors in SpaceX, which is preparing to file the largest initial public offering in Wall Street history.
In letter In a letter to investors published Tuesday, former employees highlighted what they described as “underpriced risks” associated with xAI that could complicate SpaceX’s reported plans to acquire $75 billion as part of the IPO. The rocket company’s private valuation soared to more than $1 trillion following its acquisition of xAI last year. Musk claimed his rocket company could launch data centers for his artificial intelligence lab into space, but the letter’s authors argue that xAI’s destitute track record on safety issues could complicate how investors view the combined company once it is ready to fold. Submission of an IPO prospectus.
One of the letter’s signatories and co-authors is a fresh nonprofit called Guidelight AI Standards, co-founded by former OpenAI security researcher Steven Adler and former OpenAI policy advisor Page Hedley. The group, backed by private donors, aims to improve the security practices of pioneering artificial intelligence companies. Other nonprofits focused on AI safety have also joined the project, including Legal Advocates for Protected Science and Technology, Encode AI and The Midas Project.
Hedley tells WIRED in an interview that he believes xAI has the worst security practices “in almost every case” compared to other pioneering AI developers, including OpenAI, Google DeepMind and Anthropic. As a result, he believes, SpaceX may be exposed to greater risk of regulation and litigation than other artificial intelligence labs.
The letter’s authors argue that SpaceX should disclose several information to investors, including whether xAI intends to continue developing pioneering artificial intelligence models. Most recently, SpaceX entered into an agreement to sell a significant portion of the computing power of its graphics processors to Anthropicthe letter said the agreement “leaves unclear whether xAI continues to be an AI pioneer within the larger holding company.” The authors argue that if xAI continues to pioneer AI models, it should be required to publish a public safety and governance plan.
SpaceX and xAI did not immediately respond to WIRED’s request for comment.
The letter also highlights examples of how xAI has lagged behind standard industry security practices, such as publishing detailed frameworks for mitigating the risks associated with the exploit of AI models in cyberattacks. The authors also describe specific security incidents in xAI that they believe require additional analysis. The most noteworthy was when xAI’s flagship AI chatbot, Grok, spontaneously raised the issue of white genocide in its responses. In another instance, xAI enabled Grok to generate thousands of sexualized images of women and children that were widely distributed on Musk’s social media platform X. In the latter case, at least 37 U.S. attorneys general sent a letter demanding that Musk’s artificial intelligence lab take steps to protect women and children on its platform.
Hedley says the number of security incidents xAI has experienced and the regulatory attention it has received are “far out of proportion to the company’s market share.” As lawmakers become increasingly concerned about the cyber capabilities of advanced artificial intelligence models like Anthropic’s Claude Mythos, fresh security regulations may be on the horizon. Apparently the Trump administration has already done this weighing executive order this would give U.S. intelligence agencies greater oversight of artificial intelligence models.
“Serious investments are needed to bring the situation under control [AI safety] risk and it appears that xAI has under-invested in this area in the past,” says Adler. The letter cites a report by The Washington Post which shows that xAI has “two or three” people who have been working on security since January. “The question investors should be asking themselves is whether xAI will remain on the frontier, how expensive it might actually be to manage these [risks] responsibly? If they don’t do this, what could be the consequences?”