Nvidia weighs $10 billion investment in Anthropic IPO as AI firm targets historic $2 trillion valuation

Nvidia Considers Historic $10 Billion Investment in Anthropic IPO

Nvidia is in discussions to invest as much as $10 billion in Anthropic's upcoming initial public offering, positioning itself as an anchor investor in what could become the largest IPO in history1

1

2

2

. Anthropic is seeking to raise up to $100 billion at a valuation approaching $2 trillion, marking a dramatic leap from its $965 billion post-money valuation achieved during a May funding round3

3

. The discussions remain preliminary and could change, according to sources familiar with the negotiations4

4

.

The potential Nvidia investment in Anthropic would give the Claude AI maker an early strategic backer, potentially boosting investor confidence in an offering that tests public-market appetite for frontier AI labs' enormous valuations and capital requirements3

3

. Anchor investors typically commit to purchasing a set portion of an IPO before broader marketing begins, providing crucial early validation. This approach has become increasingly common for mega IPOs, with examples including Arm's listing where Nvidia and Amazon served as anchor investors3

3

.

Anthropic's Explosive Revenue Growth Fuels IPO Ambitions

Anthropic's financial trajectory justifies its ambitious valuation targets. The company's annualized revenue run rate surpassed $65 billion by the end of July, a dramatic increase from roughly $9 billion at the end of 20253

3

4

4

. Earlier reports indicated that Anthropic's quarterly revenue passed $11.5 billion in August, representing a fourteen-fold increase year-over-year1

1

. The company is basing part of its valuation expectations on projections of approximately $190 billion to $200 billion in revenue by 20283

3

4

4

.

The listing is expected to complete before the U.S. midterm elections in November, adding a tight timeline to the ambitious fundraising effort3

3

4

4

. If successful, the raise would dwarf SpaceX's $75 billion raise in June at a $1.77 trillion valuation and exceed Porsche's EUR 9.4 billion listing in 2022, Europe's largest in decades1

1

.

Complex Web of AI Infrastructure Dependencies

The deepening relationship between Nvidia and Anthropic highlights the intricate ties between AI model developers and computing power suppliers. Anthropic relies heavily on Nvidia GPUs for AI model training and inference while simultaneously working to diversify its chip suppliers as demand for Claude strains available computing capacity3

3

. In November 2025, Nvidia committed to invest up to $10 billion in Anthropic as part of a broader partnership where Anthropic agreed to purchase $30 billion of Microsoft Azure computing capacity powered by Nvidia chips3

3

.

Anthropic already counts major tech companies as backers and compute suppliers. In April, the company committed more than $100 billion over a decade to Amazon AWS while planning to use more than 1 million Trainium2 chips3

3

4

4

. The company has also secured agreements with Google and Broadcom to add multiple gigawatts of TPU capacity3

3

. To gain greater control over AI hardware costs, Anthropic is forming an in-house team to design custom AI chips tailored specifically for Claude3

3

.

Nvidia's Strategic Pattern Across AI Investment Landscape

Nvidia's potential $10 billion commitment to Anthropic follows a pattern of strategic investments in AI companies that purchase its chips. Just three days before the Anthropic news emerged, Nvidia returned as an investor in Mistral AI's EUR 3 billion round, which closed on September 8 at a valuation above EUR 21 billion1

1

. That round represented the largest equity raise a European technology company has ever completed. The contrast is striking: Nvidia's potential cheque for Anthropic exceeds Mistral's entire fundraise and approaches half of Mistral AI's total valuation1

1

.

This investment approach has drawn scrutiny. The Bank for International Settlements has warned that the terms of arrangements where Nvidia invests in companies that buy its chips, supplies them hardware, and in some cases leases equipment back, are poorly disclosed1

1

. The pattern extends beyond direct investments. Nvidia recently agreed to acquire Hugging Face for $12.9 billion, extending its reach into software and tools developers use to build artificial intelligence2

2

. Hugging Face serves more than 18 million developers and hosts over three million models, according to Nvidia CEO Jensen Huang2

2

.

Geopolitical Implications and Market Power Dynamics

The scale difference between Nvidia's investments in Anthropic versus Mistral AI reveals how the chipmaker ranks AI companies across geopolitical lines. Mistral's EUR 3 billion round funds compute infrastructure inside Europe, targeting a gigawatt of capacity by 2030, with a business model built on selling AI sovereignty to governments and regulated industries1

1

. French President Emmanuel Macron has characterized this approach as a third way in AI, distinct from U.S. and Chinese models1

1

.

European institutions will likely be offered participation in the Anthropic IPO, priced in part by an American supplier that already owns a stake in the alternative European approach1

1

. This creates complex dynamics where cloud computing providers and chip suppliers simultaneously back competing AI model developers while those same AI companies depend on their infrastructure. As demand for AI model training and inference continues accelerating, control over compute infrastructure becomes increasingly strategic, raising questions about market concentration and geopolitical competition in the AI sector.