Nvidia's £100m-Plus AI Startup Investments Revealed: What It Means for Europe

The £100 Million Club: Nvidia's Mega-Investments

Nvidia's portfolio reveals calculated bets on foundational AI technologies, large language models, and critical infrastructure. Many involve funding rounds exceeding £100 million, cementing relationships that guarantee future hardware sales.

OpenAI received Nvidia's first direct investment in October 2024, contributing £100 million to a massive £6.6 billion funding round. While modest compared to other backers, it marked Nvidia's formal entry into funding the company that triggered the current AI frenzy. More striking was the November 2025 commitment of £10 billion to rival AI lab Anthropic.

"This is a dream come true for us. We've admired the work of Anthropic and Dario for a long time," said Jensen Huang, Nvidia's chief executive, at the time of the Anthropic announcement.

The Anthropic deal exemplifies Nvidia's circular investment logic. Anthropic committed to spending £30 billion on Microsoft Azure compute capacity whilst purchasing Nvidia's upcoming Grace Blackwell and Vera Rubin systems. This symbiotic relationship guarantees hardware demand whilst fostering the model innovation that drives further chip adoption.

Cursor, the AI-powered coding assistant, attracted Nvidia's strategic backing as part of a £2.3 billion Series D round in November 2025. The deal valued Cursor at £29.3 billion, a fifteen-fold increase within a single year.

Building the AI Infrastructure Ecosystem

Nvidia's investments extend well beyond headline-grabbing AI labs into the infrastructure companies that power them. Crusoe and Nscale, both critical partners in OpenAI's ambitious Stargate project, received substantial backing through Crusoe's £1.4 billion Series E round and Nscale's £433 million funding. Nscale, notably, operates data centres in Norway and the Netherlands, giving it a direct footprint in European AI infrastructure.

The company also backed xAI, Elon Musk's AI venture, a decision Huang defended enthusiastically. "The only regret is that I didn't give him more money. Almost everything that Elon is a part of, you really want to be a part of as well," Huang told CNBC in October 2025.

European competition authorities are paying close attention to precisely this kind of inter-locking investment web. Margrethe Vestager, the former European Commission Executive Vice President for digital policy, repeatedly signalled before leaving office that the Commission would scrutinise strategic investments by dominant technology platform companies for signs of market foreclosure. Her successor in the competition brief, Teresa Ribera, has indicated that the Digital Markets Act toolkit is available to examine whether such ecosystem-building by designated gatekeepers distorts competitive dynamics.

Strategic Scope: From Healthcare to Robotics

Nvidia's investment thesis encompasses the entire AI value chain. The company backed Scale AI with £1 billion in May 2024, securing a relationship with a crucial data-labelling provider essential for training AI models. Healthcare AI attracted attention through Hippocratic AI, which develops large language models for medical applications and received Nvidia's backing in a £141 million Series B round.

The robotics sector features prominently, with investments in humanoid robotics firm Figure AI through a £1 billion Series C round, and in autonomous driving startup Wayve via a £1.05 billion funding round. Wayve is headquartered in London, making it one of the few portfolio companies with a direct European base, and its inclusion underscores that Nvidia's ecosystem-building ambitions reach into the EU and UK markets explicitly.

For European AI labs, the dynamics are instructive. Mistral AI, the Paris-based large language model developer and one of Europe's most closely watched frontier AI companies, has so far charted an independent course, raising capital primarily from European and US venture funds without taking Nvidia's money directly. That independence may become harder to maintain as compute costs rise and Nvidia's circular investment model tightens its grip on the global supply chain.

Researchers at ETH Zurich, which operates some of the most advanced AI compute clusters in continental Europe, have noted that hardware access is already a binding constraint for academic labs. The concern is not merely academic: if the dominant chip supplier also funds the dominant model companies and the dominant cloud providers, the resulting concentration could narrow the range of AI architectures that receive serious investment.

Regulatory Scrutiny Is Intensifying

Nvidia's strategy faces growing regulatory pressure on both sides of the Atlantic. The UK's Competition and Markets Authority has been conducting a broad review of AI foundation model markets, with a specific focus on whether vertical integration and strategic investment by large incumbents risk locking in particular technology stacks. The CMA's interim reports have stopped short of enforcement action against Nvidia specifically, but the framework for intervention is being constructed.

In Brussels, the European AI Office, established under the EU AI Act, is focused primarily on model risk classification rather than investment patterns. However, the Commission's merger control and antitrust units retain full competence to act if Nvidia's investment activity is found to amount to market foreclosure or to raise barriers to entry for European competitors.

What This Means for European AI Stakeholders

For European startups, Nvidia's investment activity is a double-edged proposition. Taking Nvidia's money accelerates access to cutting-edge hardware, opens doors to the company's vast partner network, and signals credibility to other investors. The trade-off is a structural dependency on Nvidia's technology roadmap and, potentially, on its commercial priorities.

Competition from alternative chip architectures, including those being developed by UK-based Graphcore (now under SoftBank ownership) and by European research initiatives funded through the European High Performance Computing Joint Undertaking, offers a partial counterweight. But none of these alternatives yet matches Nvidia's scale, software ecosystem, or investment firepower.

Nvidia's strategy is, ultimately, ecosystem engineering at unprecedented scale. By systematically backing companies across the AI value chain, the company is not merely selling hardware; it is architecting the conditions under which AI develops globally. European regulators, founders, and policymakers would be unwise to treat this as a distant American story. The portfolio companies are building products used in European enterprises, training models on European data, and competing directly with European AI developers. The decisions being made in Santa Clara today will shape the competitive landscape that European AI companies inhabit for the next decade.