NVIDIA, Ineffable Intelligence Team Up to Build the Future of Reinforcement Learning Infrastructure
Summary
NVIDIA and David Silver's Ineffable Intelligence have partnered to build the infrastructure for large-scale reinforcement learning, focusing on pipelines that generate data on the fly and leverage NVIDIA's next-generation platforms.
<div id="bsf_rt_marker"></div><p><span style="font-weight: 400;">Reinforcement-learning agents — AI systems that learn by trial and error — can convert computation into new knowledge.</span></p>
<p><span style="font-weight: 400;">That’s the focus of a new engineering-level collaboration between NVIDIA and Ineffable Intelligence, the London-based AI lab founded by AlphaGo architect David Silver in the wake of Ineffable’s emergence from stealth last week.</span></p>
<p><span style="font-weight: 400;">“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.”</span></p>
<p><span style="font-weight: 400;">Silver is one of the pioneers of reinforcement learning, an approach that has transformed AI research. He’s focused on further developing this approach into a new paradigm.</span></p>
<p><span style="font-weight: 400;">“Researchers have largely solved the easier problem of AI: how to build systems that know all the things humans already know,” Silver said. “But now we need to solve the harder problem of AI: how to build systems that discover new knowledge for themselves. That requires a very different approach — systems that learn from experience.”</span></p>
<p><span style="font-weight: 400;">That kind of learning needs a powerful and highly optimized pipeline to support it. </span><span style="font-weight: 400;">U</span><span style="font-weight: 400;">nlike pretraining, where a fixed dataset of human data flows through the system, reinforcement learning workloads generate their data on the fly. </span></p>
<p><span style="font-weight: 400;">The system has to act, observe, score and update continuously in tight</span> <span style="font-weight: 400;">loops, which puts pressure on interconnect, memory bandwidth and serving in ways that</span><span style="font-weight: 400;"> pretraining doesn’t. Furthermore, the system will train on rich forms of experience that are quite distinct from human language and other human data, and may require novel model architectures and training algorithms. </span></p>
<p><span style="font-weight: 400;">That’s where NVIDIA and Ineffable are focusing their technical work: building a pipeline</span><span style="font-weight: 400;"> that can feed reinforcement learning systems at scale. </span><span style="font-weight: 400;">Engineers from both companies have teamed up to explore </span><span style="font-weight: 400;">the best way to create this training pipeline. </span></p>
<p><span style="font-weight: 400;">This work is starting on NVIDIA Grace Blackwell, and will be among the first to explore the upcoming NVIDIA Vera Rubin platform. The goal is to understand the next generation of hardware and software that will be required as the AI world shifts beyond human data toward models that learn through simulation and experience. </span></p>
<p><span style="font-weight: 400;">Getting this infrastructure right will unlock an unprecedented scale of reinforcement learning in highly complex and rich environments, allowing agents to discover breakthroughs across all fields of knowledge. </span></p>
View Cached Full Text
Cached at: 05/13/26, 06:12 PM
# NVIDIA, Ineffable Intelligence Team Up to Build the Future of Reinforcement Learning Infrastructure
Source: [https://blogs.nvidia.com/blog/ineffable-intelligence-reinforcement-learning-infrastructure/](https://blogs.nvidia.com/blog/ineffable-intelligence-reinforcement-learning-infrastructure/)
Reinforcement\-learning agents — AI systems that learn by trial and error — can convert computation into new knowledge\.
That’s the focus of a new engineering\-level collaboration between NVIDIA and Ineffable Intelligence, the London\-based AI lab founded by AlphaGo architect David Silver in the wake of Ineffable’s emergence from stealth last week\.
“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\.”
Silver is one of the pioneers of reinforcement learning, an approach that has transformed AI research\. He’s focused on further developing this approach into a new paradigm\.
“Researchers have largely solved the easier problem of AI: how to build systems that know all the things humans already know,” Silver said\. “But now we need to solve the harder problem of AI: how to build systems that discover new knowledge for themselves\. That requires a very different approach — systems that learn from experience\.”
That kind of learning needs a powerful and highly optimized pipeline to support it\.Unlike pretraining, where a fixed dataset of human data flows through the system, reinforcement learning workloads generate their data on the fly\.
The system has to act, observe, score and update continuously in tightloops, which puts pressure on interconnect, memory bandwidth and serving in ways thatpretraining doesn’t\. Furthermore, the system will train on rich forms of experience that are quite distinct from human language and other human data, and may require novel model architectures and training algorithms\.
That’s where NVIDIA and Ineffable are focusing their technical work: building a pipelinethat can feed reinforcement learning systems at scale\.Engineers from both companies have teamed up to explorethe best way to create this training pipeline\.
This work is starting on NVIDIA Grace Blackwell, and will be among the first to explore the upcoming NVIDIA Vera Rubin platform\. The goal is to understand the next generation of hardware and software that will be required as the AI world shifts beyond human data toward models that learn through simulation and experience\.
Getting this infrastructure right will unlock an unprecedented scale of reinforcement learning in highly complex and rich environments, allowing agents to discover breakthroughs across all fields of knowledge\.
Similar Articles
Nvidia's Jensen Huang bets on this British startup to build 'next frontier' of AI (3 minute read)
Nvidia announces a partnership with Ineffable Intelligence, a British AI startup founded by former DeepMind scientist David Silver, to develop reinforcement learning systems for superintelligence. The startup raised a record $1.1 billion seed round.
@CNET: From the Nvidia GTC Keynote, CEO Jensen Huang talks about the inference inflection point we're at.
NVIDIA CEO Jensen Huang highlighted an inflection point in AI inference during the GTC keynote, while Supermicro is partnering with NVIDIA to deliver turnkey 'AI Factory' infrastructure solutions built around the Blackwell platform.
@rohanpaul_ai: Quite a massive inferencing rack breakthrough from @TensordyneInc . They just announced an AI-inference rack, claiming …
Tensordyne announces the Napier AI inference rack, claiming 13x the throughput of Nvidia's NVL72 GB300 by using log-space math to reduce energy and transistor usage, potentially disrupting the inference hardware landscape.
NVIDIA Enables the Next Era Of Physical AI Research With Agent Skills For Autonomous Vehicles, Robotics And Vision AI
NVIDIA announced new physical AI agent skills at CVPR to accelerate research in autonomous vehicles, robotics, and vision AI, including tools for neural reconstruction, simulation, and reinforcement learning.
NVIDIA's AI agents taught robots to install GPUs into motherboards without any human help
NVIDIA's ENPIRE framework, developed with CMU and UC Berkeley, uses AI coding agents to autonomously train robots for high-precision physical tasks like GPU installation, achieving a 99% success rate through a closed feedback loop and real hardware trials.