Elon Musk’s artificial intelligence startup, xAI, has claimed a strong foothold in the billion-dollar AI market, leading to a highly elevated valuation of Musk’s companies. However, xAI’s “Grok” chatbot has been a complete failure within the U.S. government sector. This fact was revealed by Reuters after reviewing government documents and conducting interviews with officials.
What do government statistics reveal? According to the 2025 federal records, there are more than 400 instances of AI usage across U.S. federal agencies. Among these, Grok AI had been utilized in only three cases. In stark contrast, OpenAI’s ChatGPT and Microsoft Copilot were used in 234 cases. Google’s Gemini was employed in 33 tasks, while Anthropic’s “Claude” was used in 26. xAI attempted to attract U.S. government agencies by pricing Grok at just 42 cents per agency, an almost nominal charge.
Technology companies often deploy such strategies initially to familiarize clients with their products. Yet despite this extremely low price, government employees have shown a clear preference against Grok. These statistics raise questions about xAI’s ambitious commercial valuation. Musk had promised to generate billions by providing AI to large corporations and institutions, but the lack of demand within the government sector has stalled those plans.
Analysts suggest that without government endorsement, private companies are also likely to harbor doubt about Grok. This could indicate shortcomings in Grok’s security features and overall quality. Musk had aggressively lobbied to increase Grok’s usage within government, conspiring with former President Donald Trump to promote widespread AI adoption across federal agencies. Musk’s Department of Government Efficiency (DOGE) also pressed for Grok’s deployment. However, the Department of Homeland Security rejected permission to use Grok. The Pentagon and several other ministries have only allowed Grok to remain at pilot stages; currently, only the Department of Energy and the Election Assistance Commission actively use it.
Within those agencies, Grok is generally employed for mundane tasks such as drafting routine letters or posting on social media. Despite a $200 million contract between xAI and the U.S. Department of Defense, staff there prefer competing AI tools. The Pentagon’s research wing utilizes Google Gemini for engineering analysis and Claude AI for coding tasks. Technical personnel do not consider Grok to be the leading market model. Recently, Grok lost a contract with the Department of Veterans Affairs because it failed to meet the agency’s specified technical standards. Grok’s performance lagging behind competitors is evident not only in the government sector but also in the private market.
A report from a technology research institution notes a consistent decline in Grok’s adoption within the corporate sector. Previously, five out of every thousand users used Grok, but that number has now dropped to just two. Additionally, Grok users spend only about half the time on the app compared to ChatGPT users. This sends a clear message that Grok AI is unlikely to penetrate the mainstream of both U.S. corporate and government markets. Meanwhile, Google, Microsoft, and OpenAI continue to lead strongly in the AI marketplace, with xAI trailing behind.
