Ineffable Intelligence is a London-based AI research lab founded in late 2025 by David Silver — former head of reinforcement learning at Google DeepMind and the scientist behind AlphaGo, AlphaZero, and MuZero. The company is building a "superlearner": an AI system that discovers all knowledge — from basic motor skills to advanced mathematics and science — entirely from its own experience, with no human-generated training data. The core mechanism is reinforcement learning, where the system improves through trial and error rather than studying internet text. The explicit goal is superintelligence: AI that surpasses human capability across every domain. There is currently no product and no revenue; the company is in pure research mode. Silver's intellectual thesis, co-developed with RL pioneer Richard Sutton, holds that large language models are fundamentally capped by their reliance on human data, and that only self-directed experience can unlock the next leap. Eventual customers would be enterprises, governments, and research institutions demanding frontier AI capabilities.