Subvolts take AI is now engineering atoms: learn how LLMs orchestrate physical experiments for new materials. Apr 3, 2026 - Developers
- Learn how AI models can be used to manage and automate complex physical experiments, potentially creating new tools for scientific discovery.
- Business
- Understand how AI is moving into physical sciences, offering new avenues for materials innovation and R&D that could lead to new products and industries.
This episode features a deep dive into a novel application of AI in materials science by a company's co-founder.
AI for Atoms: How Periodic Labs is Revolutionizing Materials Engineering with Co-Founder Liam Fedus
Audio streams from Conviction. Your place is remembered on this device.
From Subvolts
Summary and analysis by Subvolts editorial, AI-assisted.
No Priors hosts Sarah Guo and Elad Gil interview Liam Fedus, co-founder of Periodic Labs. Fedus discusses applying AI scaling laws to materials engineering, moving from dark matter physics to AI research at Google Brain and OpenAI. Periodic Labs connects large language models to the physical world to solve data challenges in materials science, using models as an orchestration layer for closed-loop experiments.
Imagine using smart computer programs, like those that understand language, to help discover new materials. This episode explains how AI is being used to speed up experiments in the real world, connecting digital intelligence with the building blocks of everything.
This episode offers insight into how AI, specifically large language models, can be applied beyond digital tasks to physical sciences like materials engineering. It's for AI researchers, material scientists, and anyone interested in the practical applications of AI in the physical world. The discussion on using LLMs as orchestration layers and connecting them to physical experiments aligns with current frontier lab efforts to bridge AI with real-world systems and automation.
- HostSarah Guo
- HostElad Gil
- GuestLiam FedusCo-Founder, Periodic Labs
- Episode Title: AI for Atoms: How Periodic Labs is Revolutionizing Materials Engineering with Co-Founder Liam Fedus
- Show: No Priors
- Length: 29 minutes
- Guest Background: Pivoted from dark matter physics research to AI, including stints at Google Brain and working on ChatGPT at OpenAI.
- Periodic Labs' Approach: Connecting massive language models to the physical world to overcome data bottlenecks in material science.
- AI Application: Using language models as an orchestration layer for closed-loop physical experiments.
What happens when you apply the scaling laws of large language models to the physical work of atoms? Elad Gil sits down with Liam Fedus, co-founder at Periodic Labs, which is pioneering an AI foundation lab for atoms. Liam discusses how he pivoted from dark matter physics research to the front lines of artificial intelligence, including stints at Google Brain and working on ChatGPT at OpenAI. He talks about how Periodic is connecting massive language models to the physical world to overcome data bottlenecks in material science. Liam also shares how they use language models as an orchestration layer operating alongside specialized neural nets to run closed-loop physical experiments. They also explore the future of AGI and ASI, as well as the role of robotics in lab automation.
Sign up for new podcasts every week. Email feedback to show@no-priors.com
Follow us on Twitter: @NoPriorsPod | @Saranormous | @EladGil | @LiamFedus | @periodiclabs
Chapters:
00:00 – Cold Open
00:05 – Liam Fedus Introduction
00:39 – Liam’s Background at Google Brain, OpenAI
05:14 – From ChatGPT to Materials and Atoms
06:34 – Training Data in the Physical World
09:52 – Generalization Across Domains
11:31 – Models as an Orchestration Layer
12:48 – Commercialization and Business Model
16:10 – How Periodic’s Success May Shape the Future
17:45 – Multidisciplinary Scaling
19:41 – Capital and Compute
21:12 – Hiring at Periodic
21:44 – Thoughts on AGI and ASI
23:30 – Timeline for Machine-Directed Self-Improvement
25:39 – Automation and Data Generation
27:59 – Why Liam is Excited About the Future of Robotics
29:25 – Conclusion
- scaling laws
- Liam Fedus
- rules showing how performance improves as size or resources increase
- orchestration layer
- Liam Fedus
- a system that manages and coordinates other systems or processes
- closed-loop physical experiments
- Liam Fedus
- experiments where results automatically trigger the next step without human intervention
39 min1 h 1136 minSame guests, other shows
1 h 01Training an AI Scientist with Feedback from Reality, w- Liam Fedus & Ekin Dogus Cubuk (from a16z)
WithLiam Fedus, Ekin Dogus Cubuk
Questions people ask
- What is Periodic Labs?
- Periodic Labs is pioneering an AI foundation lab for atoms, connecting large language models to the physical world to overcome data bottlenecks in material science.
- How does Periodic Labs use AI?
- They use language models as an orchestration layer operating alongside specialized neural nets to run closed-loop physical experiments, applying AI scaling laws to materials engineering.
Page generated Oct 19, 2026. Summaries are Subvolts' own; the audio, artwork and notes belong to Conviction.