British venture capital funding is booming – if you are an AI firm

This article first appeared as part of the Boardroom Sensemaker newsletter – The Observer’s weekly newsletter for the board chairs, directors and executives in charge of Britain’s biggest businesses. To receive it in your inbox, featuring content exclusive to the newsletter, sign up here.

British venture capital funding hit a record high of $17bn in the first half this year, with nearly three quarters going to AI companies. But total early stage investments, pre-seed and seed, both dropped last year, to $211m and $1.2bn respectively.

So what? “Software is eating the world,” quipped venture capitalist Marc Andreessen in 2011. The same is now true of AI. Dozens of listed companies, from former shoe makers (SmartBird) to golfing services and gold miners (BluSkyAI), have made the painful pivot into AI businesses, with mixed success (both have seen their stock whipsaw since rebranding).

AI-adjacency is becoming a must for founders raising capital too; it’s a brave soul who dares to omit the word “agentic” from their pitch deck. Investors are, perhaps rightly, wary of betting on anyone who ignores the industrial revolution taking place. But if AI eats up the capital, who starves?

Fighting for scraps. Figures recently published by the British Business Bank show that

• •

  while AI-related investment rose 48% last year in the UK,

• •

  seed-stage deals dropped 27% and

• •

  university spinout equity deals dropped by a third and spinout investment halved.

The latter is particularly concerning given the emphasis the UK government and others, the BBB included, have put on the country’s strength in commercialising cutting-edge research. It has led to fears that vital, future-proofing innovations in healthcare, climate and consumer tech are being left to wither on the vine.

“It’s right that we strive to become globally competitive in AI, but it really shouldn’t happen at the cost of the pipeline that sits beneath that,” says Pan Demetriou, expert in residence for VC Investment at Imperial College London. “If capital becomes too concentrated, we’ll ultimately fund a narrower range of solutions to the problems we’re increasingly facing.”

Time for a correction? In March, Yale University’s endowment disclosed it had cut venture exposure to AI from 35% to 22% of its total VC allocation, rotating into “hardware” like robotics and manufacturing instead. Will others soon follow?

“VCs are people with a hammer seeing the whole world’s problems as AI nails,” notes Cole Lundquist, a principal at QED Investors. “Though a modicum of skepticism has snuck back… without a core AI story it’s nigh-on impossible for a company to get funded.” This has led to what he terms “malinvestment” at the earliest high-risk stages, where funds that previously would make long-shot deals crowd into “also-ran” AI companies.

…or more concentration? Data from the US’s National Venture Capital Association and Pitchbook shows that institutional investors directed 91% of their fund commitments in the first quarter of 2026 to “brand-name” VC firms – such a16z or Sequoia – rather than smaller managers writing smaller seed cheques. “Funds are feeling the pressure to start showing distributions [cash back from capital paid in] – and the earlier you go into a company, the longer you have to wait for that,” says Demetriou.

The human cost. It’s real. A survey of 165 founders by the VC and accelerator Unrest found that

• •

  1 in 3 are fundraising with six months’ runway or less, and 13% have under three months of cash remaining;

• •

  29% say they have experienced some kind of pressure from investors to include AI in their product, roadmap or pitch; and

• •

  28% admit hiding serious financial pressure from family, co-founders or colleagues during fundraising, while 22% have used personal savings to keep their business going.

This is not all due to AI, of course. Much of it has to do with the UK’s patchy safety net for entrepreneurs. Indeed, in at least one way, AI has made the job of a founder more simple.

“It forces founders to strip down what is the value that they, their brains, are adding to make this happen, because you can basically vibe-code a tool in 15 minutes,” says Gabrielė Barteškaitė, co-founder of Future Greens, a Sheffield-based biowaste startup.

What’s more… of the roughly $300bn that startups raised globally in the first quarter of 2026, $188bn went to just four companies: OpenAI, Anthropic, xAI and Waymo.

Photograph by Bloomberg via Getty Images