Nvidia CEO Jensen Huang told attendees at the Morgan Stanley Tech, Media and Telecom conference in downtown San Francisco on Wednesday, that his company’s recent investments in OpenAI and Anthropic are likely to be its last in both firms.
Huang indicated the window for such private investments typically closes once companies go public, which both AI labs are widely expected to do later this year. The remarks came during a wide-ranging discussion on Nvidia’s ecosystem strategy. Huang emphasized that the company’s stakes in OpenAI and Anthropic were made “very squarely, strategically” to expand and deepen Nvidia’s reach in the AI ecosystem — a goal he said has largely been achieved.
Huang had reiterated during the company’s fourth-quarter earnings call that all Nvidia investments focus on ecosystem expansion rather than purely financial returns. This means that Nvidia does not need additional upside from equity stakes in either company to benefit from their growth. The company remains the dominant supplier of GPUs that power both OpenAI’s ChatGPT and Anthropic’s Claude models, generating massive recurring revenue from data center sales.
Register for Tekedia Mini-MBA edition 19 (Feb 9 – May 2, 2026).
Register for Tekedia AI in Business Masterclass.
Join Tekedia Capital Syndicate and co-invest in great global startups.
Register for Tekedia AI Lab.
In that sense, the investments were more about strategic alignment and influence than financial engineering. Several factors appear to be driving Nvidia’s apparent decision to step back from further commitments. The most straightforward explanation is timing: late-stage private investments often taper off as companies approach IPOs, when public-market liquidity and valuation mechanisms take over.
Both OpenAI and Anthropic have been preparing for public listings in 2026, with OpenAI reportedly targeting a valuation north of $840 billion and Anthropic exploring a range above $360 billion. However, the relationship dynamics have grown increasingly complex. Nvidia’s $30 billion investment in OpenAI’s latest $110 billion round (finalized last week) came in well below the $100 billion figure floated in September 2025.
Huang has dismissed suggestions of bad blood between the companies as “nonsense,” but the reduction suggests a recalibration of exposure. The situation with Anthropic has been more visibly strained. In November 2025, Nvidia committed $10 billion to Anthropic. Just two months later, Anthropic CEO Dario Amodei — speaking at Davos without naming Nvidia directly — compared U.S. chip companies selling high-performance AI processors to approved Chinese customers to “selling nuclear weapons to North Korea.”
The remark drew sharp attention in Washington and within Nvidia’s orbit. The tension escalated dramatically last week when the Trump administration barred federal agencies and defense contractors from using Anthropic’s technology after the company refused to remove restrictions on mass domestic surveillance and fully autonomous lethal weapons.
Defense Secretary Pete Hegseth designated Anthropic a “supply-chain risk” and threatened broader consequences. Within hours, OpenAI announced its own Pentagon deal — a move Anthropic publicly called “mendacious.” Claude’s iOS app overtook ChatGPT as the top free app on Apple’s U.S. store in the immediate aftermath, reflecting consumer backlash.
Nvidia now holds equity in two companies that are diverging sharply in their approach to government partnerships. OpenAI has embraced defense contracts with explicit safeguards; Anthropic has drawn hard red lines and faced retaliation. This places Nvidia in an awkward position as both a major investor and the primary hardware supplier for two increasingly polarized AI labs.
Industry observers note that late-stage private investing often continues until the eve of an IPO when companies seek to maximize valuations and liquidity. Huang’s statement that the “opportunity to invest closes” with an IPO is technically accurate but somewhat simplified; many firms continue buying shares in pre-IPO windows or secondary markets.
The more likely explanation is that Nvidia is quietly stepping back from a situation that has become politically and reputationally complicated far faster than anticipated. The broader context is Nvidia’s extraordinary position in the AI boom. The company’s data center revenue — driven overwhelmingly by demand from OpenAI, Anthropic, Google, Microsoft, Meta, and others — has made it one of the most valuable companies in history.
Equity stakes in AI labs were strategic, not essential for financial upside. With both OpenAI and Anthropic heading toward public markets, Nvidia can maintain influence through GPU sales without additional balance-sheet exposure. The timing of Huang’s remarks also coincides with Nvidia’s own investor scrutiny. Shares have been volatile in recent months amid concerns over AI spending sustainability, U.S.-China tensions, and competition from emerging Chinese AI chipmakers. Signaling a pullback from further high-profile AI equity bets may reassure investors focused on Nvidia’s core semiconductor dominance rather than venture-style side plays.
Whether Huang foresaw the full political and competitive complexity of backing two leading AI labs is impossible to know. What is clear is that Nvidia’s web of partnerships has become more tangled than ever. OpenAI is deepening government ties; Anthropic is publicly defying them. Nvidia, as both investor and indispensable supplier, sits at the center of that tension.