On April 27, 2026, Bloomberg reported — and TechCrunch confirmed — that Ineffable Intelligence, the British AI lab quietly founded a few months earlier by David Silver, the former head of DeepMind’s reinforcement learning team and a University College London professor, had raised $1.1 billion at a $5.1 billion valuation, clearing the “pentacorn” threshold of $5 billion and up. Sequoia and Lightspeed led the round, with participation from Index Ventures, Google, Nvidia, the British Business Bank, and the UK’s new sovereign AI venture fund.
What makes the round unusual is what it buys: no product, no revenue — a research direction. Ineffable wants to build a “superlearner” that acquires knowledge and skills purely through reinforcement learning, by trial and error, with no human-generated training data at all. At a moment when frontier models are constrained by the shrinking supply of high-quality human data, this is the most direct bet yet on a post-human-data era.
From AlphaZero to the Superlearner
Silver’s record is the best advertisement for the route. He led DeepMind’s reinforcement learning team for more than a decade, and his most famous result is AlphaZero: starting from pure self-play, without studying a single human game, it mastered chess and Go and beat the strongest specialized engines of its day.
Ineffable’s ambition is to push the same self-learning principle off the closed board and into the open world. The company’s site compares the goal to Darwin: “his law explained all Life” and “ours will explain and build all Intelligence.” In a Wired interview, Silver described the venture as the extension of his life’s work, and pledged that any money he makes from Ineffable will go to high-impact charities that save as many lives as possible.
The Terms Behind the Round
- $1.1 billion raised at a $5.1 billion valuation, putting the months-old company straight into the $5B-plus “pentacorn” club
- Co-led by Sequoia Capital and Lightspeed Venture Partners
- Follow-on investors include Index Ventures, Google, Nvidia, plus the state-backed British Business Bank and the UK’s sovereign AI venture fund
- Several former DeepMind staff have joined Ineffable’s executive team
Note the official money in the cap table. The sovereign fund’s direct participation signals that this is not just a venture deal — it is also Britain placing national chips on a homegrown star researcher.
The Coconut-Round Era: Seed Inflation for Star Researchers
Ineffable is not an isolated case. Bloomberg has taken to calling the 2026 phenomenon “coconut rounds”: top researchers who start companies now raise Series C-scale money in their first round:
- AMI Labs, co-founded by Turing Award winner and former Meta chief AI scientist Yann LeCun, raised $1.03 billion in March 2026 at a $3.5 billion pre-money valuation
- Recursive Superintelligence, co-founded by former DeepMind principal scientist Tim Rocktäschel, reportedly closed $500 million, with demand that could have pushed it to $1 billion
Against that backdrop, Ineffable’s $1.1 billion sets another record. Capital concentrating in a handful of people whose names double as technical roadmaps suggests investors are buying options on a paradigm shift.
Why London Is the Third AI Hub
The round also reconfirms London’s standing. Google has built the city into a core AI base since acquiring DeepMind in 2014; Jeff Bezos’s Project Prometheus is reportedly scouting office space near Google’s London AI hub; and now Ineffable has gathered a bench of former DeepMind leaders under one startup roof. Talent, private capital, and state support stacking up in a single city is a combination rarely seen outside the San Francisco Bay Area and Beijing.
What It Means for Builders and Researchers
Three things to watch. First, human data stops being the only moat: if the superlearner route works, data licensing and web crawling matter less, and the capacity to generate one’s own experience matters more. Second, evaluation needs a rethink — how do you benchmark a model that never saw human knowledge? Most existing benchmarks quietly assume the boundary of what humans know. Third, zooming out, this migration of capital toward star researchers with giant seeds echoes what we sketched in our 2026 opening outlook: this year’s story is no longer scaling the existing paradigm, but finding the next one.
Sources
- DeepMind’s David Silver just raised $1.1B to build an AI that learns without human data — TechCrunch
- David Silver, Ineffable Intelligence, Reinforcement Learning — Wired
- Sequoia and Nvidia Back Ex-DeepMind Researcher at $5.1 Billion Value — Bloomberg
AI-assisted summary compiled from the sources above, reviewed by a human before publishing.