← All stories
AI & Tech

AI Labs Bet Millions of Tasks Across RL Environments Will Create AGI

Dwarkesh Patel Podcast · The next big breakthrough will be AIs learning on the job · June 26, 2026
AI Labs Bet Millions of Tasks Across RL Environments Will Create AGI
Dwarkesh Patel Podcast
Dwarkesh Patel Podcast
The next big breakthrough will be AIs learning on the job
"They think that if we train AIs to accomplish millions of verifiable tasks across thousands of diverse RL environments, then we will have basically built AGI, because this kind of training will have created a kind of problem-solving agent, the kind of thing that can make progress on open-ended tasks for weeks on end in the face of errors and mistakes and ambiguity."
Major AI research labs are making a central bet that scaling reinforcement learning across massive numbers of verifiable tasks will produce artificial general intelligence. This represents a fundamental strategic direction for companies investing billions in AI development, with the belief that current limitations in data efficiency and continual learning can be overcome through sheer compute scaling.

About this episode

In this monologue episode, AI researcher and podcast host Dwarkesh Patel examines the fundamental strategic bet major AI labs are making: that training models on millions of verifiable tasks across thousands of reinforcement learning environments will create artificial general intelligence. Patel reveals that current models are one-millionth as sample efficient as humans during training, though labs argue this inefficiency is a one-time cost amortized across billions of deployment sessions. He identifies an underrated bottleneck in AI progress: computer use capabilities lag because training requires replayable simulators, and companies like Amazon block bot training on real websites, forcing labs to build labor-intensive application clones. Patel argues that critical real-world skills like building businesses, winning elections, or succeeding in markets cannot be trained through current RL methods because they require months of real-world interaction that cannot be simulated in data centers. He cites a revealing quote from Anthropic CEO Dario Amodei suggesting short-horizon RL training may not generalize to long-horizon performance, potentially undermining the core AGI scaling hypothesis. The episode explores why continual learning and sample efficiency are deeply connected problems, discussing architectural innovations and alternative training methods like on-policy self-distillation and speculative "dreaming" approaches where AIs build and train against self-generated simulations. Patel concludes with a 2027-2028 scenario where deployed AIs learn primarily from real-world interactions across users rather than pre-deployment training, fundamentally changing how AI capabilities improve.

Key takeaways

More stories More from Dwarkesh Patel Podcast