SpaceX IPO Ambitions Face Questions As Grok Struggles In US Government
SpaceX’s planned initial public offering is expected to become the largest IPO in history, driven partly by expectations surrounding its artificial intelligence subsidiary, xAI. However, xAI’s Grok chatbot has struggled to gain traction within one of the world’s biggest potential customers — the United States government.
According to federal AI inventory records reviewed by Reuters, alongside interviews with federal employees and contracting experts, Grok has seen only limited adoption across government agencies despite aggressive promotion from SpaceX chief executive Elon Musk.
Grok Trails Rivals In Federal AI Adoption
The 2025 consolidated AI inventory from US federal agencies identified more than 400 publicly disclosed AI use cases naming specific vendors.
Only three of those examples involved xAI or Grok. In contrast, 234 use cases involved technology based on models from OpenAI, including ChatGPT, Codex and Microsoft Copilot. Another 33 involved products from Google or its Gemini models, while 26 referenced Claude from Anthropic.
Grok has reportedly been available to federal agencies for eight months at a symbolic cost of 42 cents per agency, a pricing strategy commonly used by technology firms to encourage adoption before pursuing larger contracts.
Valerie Wirtschafter, a researcher at the Brookings Institution, said such pricing models are designed to make generative AI tools essential to day-to-day government work over time.
Concerns Over Market Position
The weak federal uptake has raised broader questions about xAI’s ability to compete with established AI leaders and support SpaceX’s reported $1.75 trillion IPO valuation.
In a recent regulatory filing, SpaceX said it expects future revenue from enterprise AI services to exceed earnings from its existing businesses, estimating the broader AI services market opportunity at $26.5 trillion.
However, some analysts believe Grok’s limited adoption among government agencies may signal deeper challenges.
Vineet Jain, chief executive of AI software company Egnyte, said the government’s lack of enthusiasm for Grok could undermine confidence among corporate customers.
He argued that limited federal validation may raise concerns about security standards and reliability for enterprise buyers considering large-scale deployment.
DOGE And Federal Promotion Efforts
Musk has repeatedly promoted Grok’s potential for government work. In September, he announced xAI’s agreement with the General Services Administration and said his team wanted to work with President Donald Trump to rapidly expand AI adoption across federal agencies.
Musk’s Department of Government Efficiency (DOGE) also encouraged agencies to use Grok. Reuters previously reported that DOGE advised officials at the Department of Homeland Security to adopt the chatbot even before it had received formal approval within the agency.
Despite those efforts, Grok’s actual government usage appears limited.
At the Office of Personnel Management and the Department of Health and Human Services, Grok was mainly used for basic administrative tasks such as drafting documents or social media content.
More advanced government AI projects also showed minimal use of Grok. Federal inventory records indicated that the chatbot was only deployed in limited pilot programmes at the Department of Energy’s Lawrence Livermore National Laboratory and the Election Assistance Commission.
Pentagon Usage Remains Limited
The federal inventory does not include classified military programmes, although the Pentagon has signed a $200 million deal with xAI.
Earlier this year, Grok was added to GenAI.mil, the military’s unclassified AI platform. xAI also became one of seven companies approved for deployment on classified Defense Department networks.
Even so, one Pentagon source told Reuters that many defence staff preferred competing AI systems over Grok.
At the Defense Advanced Research Projects Agency (DARPA), engineers reportedly favour Gemini for technical analysis and Claude for coding, research and writing tasks. OpenAI systems are also used extensively, while Grok has seen little adoption.
The source added that many researchers considered Grok less capable than rival models in sophisticated engineering environments.
Corporate Adoption Challenges Continue
xAI is still attempting to expand its federal presence and is currently pursuing FedRAMP High Authorisation with support from the US Department of Agriculture.
However, USDA technology professionals interviewed by Reuters said they were largely unaware of Grok being actively used inside the department.
Last month, xAI also reportedly lost a bid to build a Grok-powered system for the Department of Veterans Affairs because the chatbot did not meet agency requirements.
Corporate adoption trends appear similarly weak. Data from web traffic monitoring company Netskope showed Grok usage among enterprise customers had fallen sharply. Updated figures shared with Reuters indicated Grok usage declined to just two users per 1,000 corporate AI users, down from a previous peak of five per 1,000.
Netskope executive Ray Canzanese said Grok users also spent significantly less time using the chatbot compared with users of ChatGPT and competing AI systems.
He added that current usage patterns suggest Grok is struggling to achieve mainstream adoption among corporate customers.
With inputs from Reuters