SpaceX AI Risks: Investors Beware

SpaceX has alerted investors to significant risks associated with its artificial intelligence features, including Grok's "Spicy" and "Unhinged" modes. These settings, designed to generate more candid outputs with reduced safety filters, could lead to regulatory scrutiny and reputational harm for the company.

As part of its planned initial public offering, SpaceX disclosed in a filing on Wednesday that it had allocated $530 million for potential litigation losses. Some of these potential losses may stem from ongoing complaints concerning sexualized imagery generated by its Grok chatbot.

The disclosures highlight the new financial and reputational risks SpaceX assumed after acquiring Elon Musk's AI startup xAI in February. This acquisition significantly boosted SpaceX's private valuation to over $1 trillion. The company repeatedly states xAI's mission is to develop "truth-seeking artificial intelligence," though this has often resulted in features launched with minimal guardrails.

SpaceX revealed it is currently under investigation in the United States and other nations. These investigations concern allegations that Grok was used to create sexualized imagery of apparent minors. The company is also a defendant in several class-action lawsuits. Future "misuse" of its AI products could lead to further regulatory sanctions, potentially including market access loss, which has occurred previously.

SpaceX noted that some AI products, like Grok's "Spicy" and "Unhinged" modes, are "designed to generate more candid, direct, or less reserved or irreverent outputs." These modes present heightened risks, including reputational damage, the generation of explicit content, misinformation, nonconsensual imagery, intellectual property infringement, and content that could be viewed as exploitative, harmful, harassing, abusive, or discriminatory.

As of March 31, Grok and X have approximately 550 million combined monthly users, with 117 million using Grok's AI features monthly. This user base is substantial, though OpenAI's ChatGPT reports over 900 million weekly users. The risks posed by Grok and X will be a key consideration for investors ahead of the SpaceX IPO.

Concerns have been raised by nonprofit groups regarding xAI's safety record potentially becoming a liability for SpaceX investors. The AI unit, including X and xAI, incurred an operating loss exceeding $6.3 billion last year. While ad, data, and subscription sales are growing, profitability is not yet imminent. A notable positive development is SpaceX's data center access deal with Anthropic, valued at $15 billion annually.