On April 28, 2026, according to Reuters, British AI unicorn Ineffable, founded by David Silver, the former chief scientist at Google DeepMind, has completed a seed – round financing of $1.1 billion (approximately RMB 7.514 billion), with a valuation reaching $5.1 billion (approximately RMB 34.831 billion). This round of financing was led by Sequoia Capital and Lightspeed Venture Partners in the United States, and co – invested by enterprises and institutions such as Nvidia, Google, and the UK’s national AI venture capital fund Sovereign AI. Ineffable declared that this round of financing is the largest seed – round financing in Europe to date.
Currently, the profit model, product launch time, and revenue scale of this startup are not yet clear.
Ineffable was founded in November 2025, with the goal of creating a Superlearner system. This system does not rely on human data and will explore all knowledge through indepfinishent practice, covering the entire range from basic motor skills to high – level ininformectual breakthroughs.
Silver, the founder and CEO of Ineffable, is best known as the core researcher behind AlphaGo. He led the research and development of AlphaGo, AlphaZero, and AlphaStar, and was fully involved in the construction and iteration of DeepMind’s reinforcement learning system.
Silver and Demis Hassabis, the co – founder and CEO of Google DeepMind, were university classmates at the University of Cambridge. During their studies at Cambridge, Hassabis taught Silver to play chess, including Go.
After obtaining a Bachelor of Arts degree from the University of Cambridge, Silver went to the University of Alberta in Canada to pursue a Ph.D. in computer science, under the tutelage of Richard Sutton, a Turing Award winner and the father of reinforcement learning.
▲David Silver (Source: Silver’s personal website)
After graduating from undergraduate studies, in 1998, Silver co – founded game company Elixir Studios with Hassabis and served as CTO and chief programmer. Later, Hassabis co – founded DeepMind with two other people. At the launchning of DeepMind’s establishment, Silver served as an advisor to the company and officially joined in 2013, staying for more than 10 years.
During his tenure at DeepMind, Silver’s research focapplyd on deep reinforcement learning, a field that combines reinforcement learning with deep learning. He participated in the research and development of several ininformigent programs. Among them, AlphaGo, led by Silver, was the first program to defeat top professional Go players in a Go game.
After that, he led the team to create AlphaZero. This program autonomously learned Go from scratch based on the same AI architecture and later mastered chess and shogi with the same training logic. Its comprehensive strength far exceeded all similar programs at the same time. In addition, he co – led the AlphaStar project. This program can reach the competitive level of human professional e – sports players in the high – difficulty strategy game “StarCraft II”.
Outside the industrial circle, Silver also serves as a professor at University College London (UCL).
At the launchning of founding Ineffable, Silver posted a personal esdeclare on the company’s blog, declareing: “The world necessarys a stage for the ambitions of the reinforcement learning paradigm to be fully realized. There, we face the fundamental proposition of ininformigence: how to (let AI) discover unknown knowledge through experience in the environment.”
He also declared: “The ecosystem for AI to generate language, video, code, etc. has been well – developed and does not necessary my involvement. Ineffable is the caapply of my life.”
According to a report by Wired yesterday, Silver declared: “All the income I obtain from the Ineffable project will be donated to charitable organizations with high social impact to save more lives.”
Conclusion:
High – value financing is concentrated in emerging AI companies
Non – large – model tracks are accelerating their breakthrough
Currently, Ineffable is still in the early R & D cycle. Its technical solution is not yet mature, and its commercialization model and implementation plan are not clear. With the support of huge capital, whether the company can break through the existing AI technology bottlenecks and balance frontier exploration with commercial sustainable development through the reinforcement learning route may become the core focus of the industest in the future.
Since the launchning of this year, the financing scale of various emerging indepfinishent AI laboratories has reached billions of dollars. The AMI laboratory, co – founded by Yann LeCun, a Turing Award winner and the former chief AI scientist at Meta, completed a $1.03 billion seed – round financing in March this year, with a pre – investment valuation of $3.5 billion.
Richard Socher, a world – class NLP scholar, is in talks for financing for his personal laboratory, and the valuation of this enterprise has reached $4 billion. In addition, Thinking Machines, an AI startup founded by Mira Murati, a former OpenAI executive, is in talks for a new round of financing, with an estimated valuation of about $50 billion.
A group of top scientific research talents have left large technology companies one after another to set up indepfinishent AI laboratories. Different from the current mainstream large – language – model track, these emerging R & D teams are breaking out of the homogeneous competition of large models and turning to frontier directions such as reinforcement learning and real – world scenario perception to explore differentiated technical routes, setting off a new wave of AI.
This article is from the WeChat official account “ZhiDongXi”, author: Liu Yu, editor: Chen Junda. Republished by 36Kr with permission.
Leave a Reply