SpaceX has issued a stark warning to potential investors, flagging the significant regulatory and reputational hazards stemming from the less-filtered, more provocative capabilities of its artificial intelligence chatbot, Grok. In a filing submitted on Wednesday as part of its planned initial public offering (IPO), the aerospace giant detailed how features like Grok’s "Spicy" and "Unhinged" modes, which are designed to produce raunchier image and voice responses with fewer safety guardrails, could expose the company to substantial legal challenges and damage its public image. This disclosure underscores the complex and often perilous landscape of advanced AI development and deployment, particularly for a company aiming to join the public markets.
The company’s proactive disclosure highlights a critical tension in the burgeoning field of generative AI: the balance between innovation and responsible development. While Elon Musk, the visionary behind both SpaceX and the AI startup xAI (which developed Grok), often champions AI that pushes boundaries and challenges conventional norms, these very characteristics have demonstrably attracted unwanted attention from regulators and civil society alike. The potential for severe financial repercussions is not merely hypothetical; as of December, SpaceX had already provisioned $530 million to cover potential litigation losses, a sum partly earmarked for ongoing complaints related to the generation of sexualized imagery by its AI unit.
The Acquisition of xAI and Escalating Risks
The financial and reputational risks associated with Grok were significantly amplified in February when SpaceX acquired Elon Musk’s artificial intelligence startup, xAI. This strategic move, which propelled SpaceX’s private valuation past the $1 trillion mark, brought the development of advanced AI directly under the SpaceX umbrella. While xAI’s stated mission, as repeatedly emphasized in SpaceX’s filing, is to develop "truth-seeking artificial intelligence," its practical application has frequently involved launching AI features with minimal inherent restrictions. This "free-wheeling nature," often presented by Musk as a distinct selling point, has placed xAI, and by extension SpaceX, in a precarious position with regulatory bodies.
Disclosing potential business risks is a standard and legally mandated component of any IPO filing. Many of the concerns outlined by SpaceX may not materialize into concrete problems. However, the company finds itself among a growing cohort of AI developers facing intense scrutiny from governments worldwide as they grapple with the profound societal implications of generative AI technologies. The ability of these tools to create realistic text, images, and audio raises complex ethical questions regarding misinformation, deepfakes, copyright infringement, and the potential for harmful or exploitative content.
Regulatory Investigations and Litigation Landscape
SpaceX’s filing explicitly reveals that the company is currently under investigation in both the United States and several other countries. These investigations are reportedly focused on allegations that Grok has been utilized to generate sexualized imagery depicting apparent minors. Such allegations represent some of the most serious potential legal and ethical breaches in the AI space, carrying immense potential for severe penalties and irreparable reputational damage.
Furthermore, SpaceX acknowledged that it is already a defendant in multiple ongoing class-action lawsuits. The company warns that any future "misuse" of its AI products could lead to additional regulatory sanctions. These sanctions could be severe, potentially including the "loss of access to certain markets, which has occurred in the past," a veiled reference to the challenges faced by technology companies operating in heavily regulated jurisdictions.
"Spicy" and "Unhinged" Modes: A Double-Edged Sword
The filing specifically calls out Grok’s "Spicy" and "Unhinged" modes as key contributors to these heightened risks. The company describes these modes as being "designed to generate more candid, direct, or less reserved or irreverent outputs." This deliberate design choice, while potentially appealing to a segment of users seeking more edgy or unfiltered AI interactions, inherently increases the probability of generating problematic content.
SpaceX’s filing elaborates on the specific dangers associated with these modes: "Because these modes may be more irreverent and harsher than our standard offerings, they present heightened risks, including reputational harm, the generation of potentially explicit content and misinformation or deceptive outputs, potential nonconsensual or exploitative imagery, intellectual property infringement, or content that could be viewed as exploitative, harmful, harassing, abusive, or discriminatory." This candid admission underscores the company’s awareness of the tightrope it is walking in offering such advanced and less constrained AI capabilities.
User Base and Competitive Landscape
Despite the disclosed risks, SpaceX also provided data on Grok and its broader platform, X (formerly Twitter), indicating a substantial user base. As of March 31, Grok and X combined boast approximately 550 million monthly users, with 117 million engaging with Grok’s AI features each month. These figures place Grok within the competitive landscape of generative AI, though still behind established players like OpenAI’s ChatGPT, which reports over 900 million weekly users. This scale of user engagement, while impressive, also amplifies the potential reach and impact of any problematic AI outputs.
The question of whether the potential rewards of these AI ventures outweigh the considerable risks and associated costs will undoubtedly be a focal point for investors considering SpaceX’s IPO. Earlier in the week, a coalition of non-profit organizations issued a warning, asserting that xAI’s documented history of safety shortcomings could indeed become a significant liability for prospective SpaceX investors. This external pressure adds another layer of complexity to the company’s public offering.
Financial Performance of the AI Division
SpaceX’s AI division, encompassing both X and xAI, has represented a significant financial undertaking, contributing to an operating loss exceeding $6.3 billion last year. While revenue streams from advertising, data, and subscriptions are showing growth, they have not yet reached a pace that would swiftly render the division profitable. This financial reality suggests that the high costs associated with AI development and infrastructure, coupled with the ongoing risks and potential liabilities, present a substantial challenge to the division’s immediate profitability.
One notable bright spot within SpaceX’s AI endeavors is its strategic partnership with Anthropic. Under an agreement valued at $15 billion annually, Anthropic will gain access to SpaceX’s data centers, providing a significant revenue stream for the company’s infrastructure services. This deal highlights the growing demand for high-performance computing power necessary for training and deploying advanced AI models, a demand SpaceX is well-positioned to meet.
Revenue Growth and Advertising Woes
The AI division’s revenue demonstrated a notable increase, reaching $3.2 billion in 2025, a rise of approximately 22 percent from the previous year. SpaceX attributed a portion of this surge to increased advertising sales on the X platform. However, the first quarter of the current year saw a significant downturn in ad sales, with a reported drop of $100 million. The company characterized this decline as a temporary issue, attributing it to an ongoing overhaul of its advertiser tools. This fluctuation in advertising revenue underscores the inherent volatility and competitive pressures within the digital advertising market, particularly for a platform undergoing significant strategic shifts.
Subscription Growth as a Stabilizing Factor
In contrast to the advertising challenges, SpaceX’s services business has shown more consistent positive performance. Revenue generated from subscriptions to Grok and X saw a substantial increase, climbing by $365 million last year and an additional $177 million in the first three months of the current year. While SpaceX has not disclosed the specific subscriber numbers for each service, the robust growth in subscription revenue suggests a growing willingness among users to pay for premium features and enhanced AI capabilities. In the United States, Grok subscriptions start at $10 per month, while X Premium is available for $3 monthly, indicating a tiered approach to monetizing its user base. This subscription-driven revenue model could offer a more stable and predictable income stream compared to the fluctuating advertising market, potentially mitigating some of the financial risks associated with the AI division.
The comprehensive disclosures made by SpaceX in its IPO filing paint a detailed picture of a company at the forefront of technological innovation, yet simultaneously navigating a complex web of ethical, legal, and financial challenges. The success of its planned public offering will hinge, in part, on how effectively it can convince investors that its ambitious AI ventures, particularly those involving the unbridled capabilities of Grok, are worth the considerable risks involved. The coming months will reveal whether the market is prepared to embrace a future shaped by AI that pushes boundaries, even as it grapples with the potential for unintended consequences.
