SpaceX is betting its AI future on Grok despite weak enterprise adoption and a scandal-ridden rollout. An S-1 filing reveals the company launched Grok Business, Grok Enterprise, Grok API, and xAI Gov, yet corporate usage barely moved from four percent to seven percent between 2025 and 2026. Meanwhile, Anthropic's Claude jumped from 21 percent to 48 percent in the same period, per Enterprise Technology Research.

Grok's low adoption comes despite xAI's aggressive push into government contracts. Reuters found just three public mentions of Grok among 400+ disclosed federal AI uses. A January 2026 update let users generate sexualized images by virtually undressing people, sparking lawsuits and an EU ban on nudifying apps. SpaceX's filing flags "Spicy" and "Unhinged" modes as carrying "heightened risks" including nonconsensual imagery and misinformation.

SpaceX acknowledges regulatory scrutiny and litigation risk in its filing. Despite Grok's tepid performance, the company remains committed to its enterprise push, leaving xAI exposed to reputational damage and monetization limits across jurisdictions.