Grok, Elon Musk's artificial intelligence chatbot, is failing to gain traction in the federal government and consumer markets, creating a significant headwind for SpaceX's planned initial public offering (IPO), which the company filed on Tuesday. Downloads have collapsed from 20 million in January to 8.3 million in April, while paid conversion rates sit at roughly 0.174% of surveyed US consumers and workers in the second quarter of 2026, compared to more than 6% for ChatGPT. The federal government's adoption of Grok has also stalled, undermining a key revenue projection that SpaceX is using to justify what would be the largest IPO in history.

Why Is Grok Losing Ground in Washington?

In September 2025, Musk signed a General Services Administration (GSA) OneGov agreement that was supposed to deliver Grok 4 and Grok 4 Fast to every federal agency at a discounted rate of $0.42 per organization for an 18-month period. This deal was positioned as a major win for xAI, Musk's artificial intelligence company, and a cornerstone of the growth narrative supporting SpaceX's IPO filing. However, the agreement has failed to convert into meaningful federal adoption, and competitors like Anthropic and OpenAI are now operationalizing their own federal-agency deals instead.

The federal stall reflects broader concerns about Grok's reliability and bias. Public Citizen has petitioned the Office of Management and Budget (OMB) twice to suspend federal use of Grok over accuracy and bias concerns, citing prior outputs the group describes as racist, antisemitic, and factually incorrect. Senator Elizabeth Warren has separately pressed Defense Secretary Pete Hegseth over the Department of Defense granting Grok classified-system access despite concerns raised by the National Security Agency (NSA) and GSA. These regulatory and political obstacles have effectively blocked the federal revenue stream that xAI was counting on.

What Is SpaceX Doing With Grok's Computing Hardware?

The most striking development is SpaceX's decision to rent out the Memphis Colossus 1 data center, a 220,000-GPU facility that was Grok's primary training environment, to Anthropic for $1.25 billion per month through May 2029. This move reveals the underlying problem: with Grok's consumer demand falling and federal adoption stalled, xAI has more computing capacity than it needs. Rather than continue investing in Grok's growth, SpaceX is monetizing the excess hardware by leasing it to Anthropic, the very company whose Claude model has been displacing Grok on federal procurement lists.

The financial picture inside SpaceX's S-1 prospectus makes this trade necessary. xAI lost $6.4 billion from operations on $3.2 billion of revenue in 2025, with revenue growth of approximately 22%, well below the published rates at OpenAI, Anthropic, and Google DeepMind. Selling compute capacity to a competitor signals to institutional investors that the underlying product cannot generate enough demand to absorb the capacity Musk originally built for it.

How Are Investors Interpreting Grok's Weakness?

SpaceX's roadshow, which is expected to begin within the next two weeks, will be the first formal test of whether institutional buyers are willing to underwrite the AI-revenue-line projection against the federal-stall and consumer-decline data that Reuters has now made public. The prospectus does not break out Grok-specific revenue from the broader xAI line, leaving institutional buyers to interpret the federal-adoption stall against the headline AI-line growth figure on their own.

The timing compounds the challenge. xAI's $420 tax-return commitment to employees has slipped past the promised payment window, and Delaware court procedural rulings against Musk in the OpenAI litigation have continued to mount. These developments, combined with Grok's visible consumer and federal decline, create a credibility gap between SpaceX's IPO narrative and the operational reality investors can now observe.

Steps to Watch the Grok Story Unfold

• S-1 Amendment Review: Watch for updated Grok-adoption disclosures in the S-1 amendment expected ahead of the roadshow launch, which would be the first formal signal of whether xAI is willing to put numbers behind the consumer and federal demand picture.
• Roadshow Investor Questions: Monitor how institutional investors respond during SpaceX's roadshow presentations, particularly questions about the Anthropic compute deal and Grok's federal stall, which will reveal whether the market views the AI-revenue projection as credible.
• Federal Adoption Tracking: Follow any announcements from federal agencies about Grok adoption or competitive selections, as these will indicate whether the GSA OneGov agreement can be salvaged or whether Anthropic and OpenAI have permanently captured the government market.
• Compute Monetization Signals: Observe whether SpaceX announces additional compute-leasing deals or capacity sales, which would further confirm that xAI cannot absorb the hardware it built for Grok's training and operation.

The core issue is straightforward: SpaceX's IPO depends on convincing public-market buyers that xAI's AI business is a growth engine, but the evidence now suggests Grok is a declining asset. Federal adoption has stalled despite a signed GSA agreement, consumer downloads have fallen 59% in four months, and paid conversion rates are a fraction of ChatGPT's. By renting out Grok's primary training infrastructure to Anthropic, SpaceX is essentially monetizing the failure of its own product rather than investing in its recovery. The next two weeks will reveal whether institutional investors view this as a pragmatic pivot or a red flag about the sustainability of xAI's revenue projections.