Nvidia bets on five UK startups. Nvidia is taking stakes in Boltz, Foresight, Hiverge, Prima Mente and Riverlane. Nvidia is planning new investments in five UK startups, its boss said last night, as it continues to spread its reach across the startup ecosystem. At an event last night in London, Nvidia CEO Jensen Huang announced the chip giant's new bets, as he also addressed questions about the West's relationship with China amid safety concerns in the AI race. Nvidia is investing in AI drug discovery startup Boltz, infrastructure predictor Foresight, AI logistics planning startup Hiverge, AI biology startup Prima Mente and quantum computing startup Riverlane. Nvidia is understood to be taking a minority stake in the five startups. Its other existing investments in the UK include Basecamp Research, Ineffable Intelligence, Synthesia and Wayve. The string of new UK investments follows just hours after Nvidia announced it was investing in four French startups, as well as making a $250m compute donation to UK AI safety and research, as it looks to underscore its commitment to the European tech ecosystem. These were French healthtech Doctolib, AI startup Entalpic, robotics startup Genesis AI and enterprise software company Pigment. Speaking outside the London event last night, Huang was asked about the West's relations with China in the AI race amid safety concerns. Huang said the West should continue to talk to Chinese AI firms. In a video recorded by a Bloomberg reporter, he said: "I have dialogue with many researchers in China, and we ought to continue to exchange dialogue. United States should do it. I advocate that the UK do that same thing. The UK is phenomenally well positioned because you have the intellectual capital, you have the position in the world." Follow the developments in the technology world. What would you like us to deliver to you?

Nvidia CEO Jensen Huang announced investments in five UK startups at a London event, backing Boltz Bio, Foresight Works, Hivergé AI, PrimaMente, and quantum computing firm Riverlane — extending Nvidia's AI-ecosystem strategy across Europe's deep-tech capital.

The geography of Quantum: what Maryland is actually building. Most states compete for quantum companies by offering incentives. Maryland appears to be doing something in addition to that. It is building the infrastructure underneath the companies so that each new arrival increases the value of the ones already there. The evidence has been accumulating for eighteen months. What looked like a series of unrelated announcements now reads as a clear strategy. The evidence. * January 2025: Governor Wes Moore proposed $27.5 million in the FY26 budget to launch the Capital of Quantum initiative, targeting $1 billion in total state, federal, and private investment over five years. IonQ was named anchor partner, with plans to grow its Maryland headquarters to 100,000 square feet and double its workforce. * April 2025: DARPA and Maryland signed a memorandum of agreement establishing the Capital of Quantum Benchmarking Hub at UMD's Applied Research Laboratory for Intelligence and Security. Each side committed up to $100 million over four years, contingent on progress. The hub's purpose is explicit: independent evaluation of whether commercial quantum approaches can achieve utility-scale performance. * September 2025: Microsoft announced a 15,000-square-foot Quantum Research Center in the Discovery District, including classified research zones, a hardware makerspace, and direct DARPA access for its Majorana topological chip. * April 2026: IQM opened its first U.S. Quantum Technology Center, focused on integrating superconducting processors with HPC providers. * May 2026: Maryland's FY27 budget allocated $20 million toward IonQ's new global headquarters, $22 million for UMD's Quantum Startup Foundry and national testbeds, $20 million for a dedicated Deep Tech Facility, and $12 million for ARLIS and quantum faculty recruitment. * June 2026: Quantum Motion established a silicon-CMOS hardware base in the Discovery District, joining the DARPA benchmarking pipeline. * September 2026: Riverlane announced its U.S. headquarters in College Park, co-locating error correction with the hardware it serves. Days later, QuEra added neutral-atom systems to the cluster. The difference. The distinction is not the number of companies. It is the links between them. A typical regional quantum strategy attracts companies with tax incentives and hopes a cluster forms. Maryland is building the shared infrastructure first - the benchmarking capacity, the talent pipeline, the federal relationships, the fabrication and testing facilities - and letting the companies plug into it. Consider what a quantum hardware company actually needs to scale: research talent, independent performance validation, error correction, control systems, federal engagement, capital, and eventually customers. Maryland has placed each of those within the same network. The DARPA benchmarking hub is the clearest example. It is not a state program with a quantum label. It is a federal evaluation mechanism that nearly twenty companies have entered, designed explicitly to separate viable approaches from hype. In a market where every architecture has its own strengths and weaknesses, credible third-party validation is scarcer than qubits. The modality diversity. The cluster now spans five hardware approaches: trapped-ion (IonQ), topological (Microsoft), superconducting (IQM), silicon-spin CMOS (Quantum Motion), and neutral-atom (QuEra). Riverlane adds real-time error correction across all of them. This matters more than diversification for its own sake. A benchmarking hub evaluating multiple modalities produces comparative data under the same set of criteria. It becomes the place where architectures are measured against each other under federal scrutiny. For the companies, that is a double-edged sword. It offers credibility if the results hold. It offers exposure if they do not. The compounding effect. The value of this cluster is not additive. It compounds. Each new company does not simply add one more tenant to a campus. It strengthens the talent pool that universities draw on, the testing infrastructure that federal agencies use, the technical community that attracts the next company, and the credibility of the region as a whole. This is difficult to build quickly and harder to replicate. Illinois has the Quantum Proving Ground. Colorado has built its own cluster around federal labs and academic research. Both are serious efforts. But the density of co-located modalities, federal benchmarking, university research, and state capital in one corridor is unusual. The things GQI is watching. Three questions will determine whether this compounds or stalls. First, does the benchmarking pipeline produce credible, differentiating performance data? If the DARPA hub cannot distinguish viable approaches from weaker ones, its value to the ecosystem collapses. Second, does the ecosystem convert into commercial deployment? Attracting research centers is not the same as producing revenue-generating quantum systems. The test is whether companies inside the cluster win contracts because of their Maryland position. Third, can another region reproduce the model? If Maryland's advantage is institutional density and federal proximity, it may be defensible. If it is simply first-mover advantage, it may not be. The strategic question. Most regional quantum strategies are recruitment strategies. They measure success by how many companies they attract. Maryland appears to be running an infrastructure strategy. It measures success by whether the pieces reinforce each other. That is a hard thing to build and a harder thing to copy. It also takes longer to show results, which is why it is easy to miss when watching announcements one at a time. Put them together, and Maryland stops looking like a collection of company moves. It starts looking like a deliberately constructed ecosystem built not around individual companies, but around the infrastructure that allows a quantum industry to scale. The obvious next question is whether this model travels. Illinois has the DARPA Quantum Proving Ground. Colorado has built its own cluster around federal labs and academic research. Over the coming weeks, Quantum Computing Report LLC intend to examine both in the same frame: what infrastructure each has assembled, how their strategies differ from Maryland's, and whether the compounding effect Quantum Computing Report LLC is describing is reproducible or whether it depends on conditions that only exist in the Washington corridor. That is the pattern worth watching. This is the first in a series examining regional quantum strategies. For deeper analysis of ecosystem formation, infrastructure investment patterns, federal program alignment, and state-level competitive positioning, GQI subscribers can access its full quantum ecosystem database and custom briefing services. To learn more, contact GQI at [email protected] or reach out directly to [email protected]. Next week, Quantum Computing Report LLC will examine Illinois, the DARPA Quantum Proving Ground, the Chicago quantum corridor, and whether a different model of ecosystem construction produces different results. September 23, 2026