Founders and Expertise
Periodic Labs was founded by Ekin Dogus Cubuk and Liam Fedus, both of whom have led groundbreaking AI research at leading institutions. Cubuk headed the materials and chemistry team at Google Brain and DeepMind, where he contributed to an AI tool that discovered over two million new crystals in a single year. Fedus served as Vice President of Research at OpenAI, playing a key role in creating the first trillion‑parameter neural network and developing advanced AI agents.
Funding and Investor Backing
The startup emerged from stealth with a $300 million seed round, attracting a roster of high‑profile investors. Backers include Andreessen Horowitz, DST, Nvidia, Accel, Elad Gil, Jeff Dean, Eric Schmidt, and Jeff Bezos. This capital infusion provides the financial foundation for building large‑scale autonomous labs and advancing the company’s ambitious scientific agenda.
Mission to Automate Scientific Discovery
Periodic Labs’ core mission is to automate scientific discovery by creating AI scientists that operate in fully autonomous laboratories. These labs will employ robots to conduct physical experiments, gather data, iterate on designs, and continuously improve through machine‑learning feedback loops. The company envisions a future where AI‑driven experimentation can rapidly generate novel materials and insights that are difficult or impossible to achieve through traditional research methods.
First Target: Next‑Generation Superconductors
The initial focus of the autonomous labs is to invent new superconductors that outperform existing materials and potentially require less energy. By automating the discovery process, the company hopes to accelerate the development of superconductors with superior performance characteristics, opening new possibilities for energy transmission, computing, and other high‑technology applications.
Generating Fresh Physical‑World Data
Beyond material discovery, Periodic Labs aims to collect extensive physical‑world data generated by its AI scientists. This data will serve as a valuable resource for training future AI models, addressing the current limitation that many scientific AI advances rely on internet‑derived datasets. By feeding AI systems with real experimental results, the company seeks to unlock new capabilities for AI across scientific domains.
Team Composition and Collaborative Experience
The startup’s team includes researchers who have contributed to notable AI projects such as OpenAI’s Operator agent, Microsoft’s MatterGen, and other AI‑driven materials science initiatives. This collective expertise positions Periodic Labs at the intersection of advanced AI research and practical scientific experimentation.
Broader Landscape of AI‑Powered Science
Periodic Labs joins a growing ecosystem of organizations exploring AI for scientific discovery. While the company brings together a distinguished team and substantial funding, it operates alongside academic efforts and other startups that are also leveraging AI to automate chemistry and materials research.
This article was written with the assistance of AI.
News Factory SEO helps you automate news content for your site.