AlphaGo architect David Silver leaves DeepMind to launch post-LLM AI startup

What’s the deal? David Silver, a senior researcher at Google DeepMind best known for leading the AlphaGo project, has left the company to found a new AI startup called Ineffable Intelligence. The London-based company aims to pursue artificial general intelligence through approaches that go beyond large language models.

The move ends a tenure of more than a decade at DeepMind, where Silver played a central role in reinforcement learning breakthroughs that helped define the lab’s reputation.

Why now? The departure comes as large language models dominate investment and research agendas across the AI industry. Silver has argued that scaling LLMs alone is unlikely to lead to superintelligence, creating space for alternative research paths.

His exit also follows a broader pattern of senior AI researchers leaving big labs to pursue more focused, independent work.

What could go wrong? Building a new AI company around non-mainstream approaches carries technical and funding risk, especially in a market captivated by LLM-centric progress. Investors may be cautious if results are slower or harder to demonstrate.

There is also the challenge of competing for talent and compute against well-capitalised incumbents such as OpenAI, AnthropicDealroom has a profile for this one. Try Dealroom →, and Google itself.

The signal: Silver’s move highlights growing unease among leading researchers about the limits of today’s dominant AI paradigm. While capital continues to flow into LLM-driven startups, parts of the research community are looking elsewhere for the next leap.

If Ineffable Intelligence succeeds, it could help legitimise post-LLM approaches and diversify where venture money and research attention flow next.

Sources:
Fortune
The Decoder

A.M.