Peter Thiel Has "Changed"! Founders Fund Shifts from "Cautious" to "Heavy Bets" on AI

Peter Thiel Has "Changed"! Founders Fund Shifts from "Cautious" to "Heavy Bets" on AI

Amid the global venture capital frenzy over AI, Founders Fund has also shifted its strategy, moving from publicly warning about an AI bubble to making large, concentrated bets on a handful of key companies.

At a recent annual investor conference, the venture capital fund Founders Fund, founded by investor Peter Thiel—known as the "Angel of Silicon Valley"—outlined its new strategy: to concentrate capital and resources on the largest AI bets in its portfolio.

According to tech media outlet Information, insiders revealed that these core targets include AI sector leader OpenAI, data center builder Crusoe, and the internally incubated nuclear fuel startup General Matter.

This decision is notable not only because Thiel himself has compared the current AI craze to the internet bubble of 1999, but also because Founders Fund has previously been relatively cautious about investing in generative AI startups. After its initial investment in OpenAI at the start of 2023, the fund did not broadly invest in AI model developers and application companies like its peers.

Now, this "less is more" approach stands in stark contrast to competitors such as Andreessen Horowitz and Lightspeed Venture Partners, who are spreading their bets widely across multiple model developers and application startups in the AI field.

These two radically different investment paths may determine which venture firms and AI companies stand out amid this capital feast.

Betting on Scale—OpenAI as the "Top Choice"

The core of Founders Fund's strategy is a belief shared with OpenAI CEO Sam Altman: The winners in AI will be those companies able to achieve massive scale the fastest.

At the investor conference, Thiel appeared onstage alongside Altman to expound this viewpoint. This strategy also explains Founders Fund’s recent investment of about $1 billion in OpenAI, which is one of the largest investments in the company’s history.

Altman stated at the meeting that OpenAI will continue to use its billions in capital to invest in advanced chips and data centers, while accelerating hiring to expand business. He said, based on an employee stock sale this week, OpenAI’s valuation has reached $500 billion, making it the world’s "highest value per employee" company, and may consider going public in the next few years when conditions are right.

It's worth noting that Thiel and Altman have a long history. Over a decade ago, Thiel invested in Altman's venture fund Hydrazine Capital; when OpenAI was founded in 2015, Thiel was also listed as one of its original donors.

Continuation of the "Contrarian" Investment Philosophy

Concentrating capital on a handful of favored companies has always been a hallmark of Founders Fund.

This concentrated betting in the AI field can be seen as a continuation of its enduring "contrarian" investment philosophy, rather than a complete strategic overhaul. This approach also aligns with Thiel's long-standing conviction that “the best investments create monopolies, not companies that simply compete in existing markets.”

In the past, the fund profited handsomely through early heavy bets on Airbnb, Palantir, and defense contractor Anduril, among others.

The classic example is SpaceX—when others hesitated to invest in rocket launches, Founders Fund decisively entered. At the conference, the fund revealed that its total investment of $671 million in SpaceX since 2008 has now grown to $20.3 billion (including current holdings and sold shares), more than 30 times return.

Now, this approach is being replicated in its latest growth fund. Founders Fund told investors it plans to back around 10 companies from its latest $4.6 billion growth-stage fund. By contrast, its smaller second and first growth funds invested in 15 and 31 companies, respectively.

The report cites a person familiar with the fund’s returns, noting that by Q1 this year, its inaugural $1.7 billion growth fund raised in 2020 achieved a 10% net internal rate of return (IRR) after fees, while the $3.4 billion second growth fund raised in 2022 achieved an IRR of 24%.

Precisely Covering the AI Industry Chain

Founders Fund’s concentrated strategy is not blind betting, but aims to support a leading company at every layer of the AI business. Through this approach, the fund tries to cover the entire supply chain, from energy and computational infrastructure to models and applications.

  • Model Layer: Founders Fund does not back OpenAI competitors like Anthropic or xAI, instead exclusively betting on OpenAI.
  • Application Layer: AI code startup Cognition is one of its largest bets. Founders Fund states Cognition may hit around $200 million in annual recurring revenue with its Devin software tool by year-end, exceeding its target. The fund recently led a $400 million funding round for Cognition, putting its valuation (including new money) at $10.2 billion.
  • Infrastructure Layer: The fund plans to participate in a new funding round for data center builder Crusoe, targeting a $10 billion valuation. Last month, Crusoe completed phase one of its Oracle data center campus in Abilene, Texas, which will provide cloud computing services for OpenAI.
  • Energy Layer: General Matter, incubated by Founders Fund and co-founded by partner Scott Nolan, is its key layout in energy. The company aims to build domestic uranium enrichment facilities in the US to supply small nuclear reactors powering AI data centers. Founders Fund has led its first two rounds and owns 27.5% of the firm.

The only field Founders Fund is currently avoiding is AI chips. Of the five AI company categories listed at the conference, chips are the only area where it hasn’t yet made a major bet; this market is currently dominated by Nvidia.

Risk Warning and DisclaimerThe market has risks, investment requires caution. This article does not constitute individual investment advice, nor does it take into account the special investment goals, financial situations, or needs of particular users. Users should consider whether any opinions, viewpoints, or conclusions in this article are suitable for their specific circumstances. If you invest based on this, you do so at your own risk.