NVIDIA Partners with David Silver's Ineffable Intelligence to Forge Next-Gen Reinforcement Learning Infrastructure

Breaking: NVIDIA and Ineffable Intelligence Announce Landmark Collaboration

NVIDIA and Ineffable Intelligence, the London-based AI lab founded by AlphaGo architect David Silver, have launched an engineering collaboration to build the infrastructure for large-scale reinforcement learning. The joint effort aims to create systems that learn continuously from experience, moving beyond static human data.

"The next frontier of AI is superlearners — systems that learn continuously from experience," said Jensen Huang, founder and CEO of NVIDIA. "We are thrilled to partner with Ineffable Intelligence to codesign the infrastructure for large-scale reinforcement learning as they push the frontier of AI and pioneer a new generation of intelligent systems."

David Silver, a pioneer of reinforcement learning, emphasized the paradigm shift: "Researchers have largely solved the easier problem of AI: how to build systems that know all the things humans already know. But now we need to solve the harder problem of AI: how to build systems that discover new knowledge for themselves."

Background: Reinforcement Learning at Scale

Reinforcement learning (RL) agents learn through trial and error, converting computation into new knowledge. Unlike pretraining on fixed datasets, RL systems generate their own data on the fly, requiring a powerful and highly optimized pipeline.

The system must act, observe, score, and update continuously in tight loops, placing unique pressure on interconnect, memory bandwidth, and serving. This work will start on NVIDIA Grace Blackwell and be among the first to explore the upcoming NVIDIA Vera Rubin platform.

Ineffable Intelligence emerged from stealth just last week, founded by Silver, who led the AlphaGo team at DeepMind. The lab focuses on developing RL into a new paradigm for discovering knowledge beyond human input.

What This Means: Unlocking Breakthroughs Across All Fields

Getting this infrastructure right will unlock an unprecedented scale of reinforcement learning in highly complex environments. Agents could discover breakthroughs across all fields of knowledge, from drug discovery to materials science and beyond.

The collaboration signals a shift from AI systems that merely mimic human data to systems that generate novel insights through experience. Engineers from both companies are exploring the best way to create a training pipeline that can feed RL systems at scale, with the goal of understanding the next generation of hardware and software required.

"That requires a very different approach — systems that learn from experience," Silver noted. The partnership aims to build the foundation for a new generation of intelligent systems that can continuously improve and adapt.