Who Is David Silver
Silver spent more than a decade leading Google DeepMind's reinforcement learning research before leaving to found Ineffable. His most notable work was AlphaZero, a system that mastered chess and Go purely through self-play without being given any human strategies, game records, or historical examples. Starting from nothing but the rules, AlphaZero defeated every existing top-performing program in each game.
That track record is the intellectual foundation for what Ineffable is attempting. Reinforcement learning — the technique by which AI systems learn through trial and error rather than from pre-existing human-labeled data — is Silver's area of expertise and the method Ineffable plans to use at scale.
The Ambition
Ineffable's stated goal is a system that learns the way AlphaZero did, but applied to knowledge and reasoning more broadly. "If successful, this will represent a scientific breakthrough of comparable magnitude to Darwin: where his law explained all Life, our law will explain and build all Intelligence," the company writes on its website.
Silver, who is also a professor at University College London, described the venture as "his life's work" in a personal note published on the company's blog. He told Wired that any personal financial gains from Ineffable would go to high-impact charities.
Part of a Larger Wave
Ineffable joins a growing cohort of well-funded AI labs founded by prominent researchers who are betting that the next major leap in AI will come from approaches different from today's dominant large language model paradigm.
Just last month, AMI Labs, co-founded by Turing Award winner and former Meta chief AI scientist Yann LeCun, raised $1.03 billion at a $3.5 billion valuation to build world models. Recursive Superintelligence, co-founded by former DeepMind principal scientist Tim Rocktäschel, has reportedly raised $500 million with demand stretching toward $1 billion. These rounds have been nicknamed "coconut rounds," a tongue-in-cheek reference to how far the concept of a startup seed round has been inflated.
What the three companies share is a conviction that the foundation model era built on massive human-generated datasets has real limits, and that the next architecturally significant systems will learn differently. Whether through world models, recursive self-improvement, or reinforcement-based discovery, these labs are making very large bets that the path forward diverges from the one that produced GPT and Claude.
London as an AI Hub
The raise also reflects a broader shift in where high-stakes AI research is being done. London is emerging as a serious alternative to the Bay Area, driven in large part by DeepMind's continued presence and the dense alumni network it has produced. Jeff Bezos' AI research lab, Project Prometheus, is reportedly in talks to establish office space near Google's London AI hub.
The UK government's participation in the Ineffable round is a signal that it views this moment as a window to anchor frontier AI development domestically. Technology minister Liz Kendall said in the government's statement: "This investment in Ineffable will support a company at the very frontier of AI, with the potential to transform entire sectors."
This article is based on reporting from TechCrunch, SiliconANGLE, and Reuters.
Image courtesy of Benjamin Davies and Unsplash.
This article was generated with AI assistance and reviewed for accuracy and quality.
