Dwarkesh Podcast Fully autonomous robots are much closer than you think – Sergey Levine
1214 snips
Sep 12, 2025 Sergey Levine, a top robotics researcher and co-founder of Physical Intelligence, believes we are on the verge of a robotic revolution by 2030. He discusses how we can pave the way for self-improving general-purpose robots that could manage our households autonomously. From the societal impacts of full automation to the challenges of scaling robotics technology, Levine emphasizes the need for proactive planning. He also explores the synergy between language models and robotics, predicting significant innovations that could transform industry and daily life.
AI Snips
Chapters
Transcript
Episode notes
Robots As General-Purpose AI Platforms
- Physical Intelligence builds robotic foundation models to control any robot for any task by leveraging general-purpose models.
- Sergey Levine argues a general robot could encompass broad AI functionality and handle many human tasks.
Home Autonomy Requires Continuous Learning
- Real household autonomy needs continuous learning, common sense, safety and long-duration responsibility, not single-use demos.
- Levine emphasizes leveraging prior knowledge and representations to scale from simple tasks to months-long responsibilities.
Data Flywheel Beats A Single Release
- The critical milestone is the start of a self-improvement flywheel where deployed robots collect experience and improve.
- Levine thinks that flywheel could begin within single-digit years and possibly very soon.