Former OpenAI and DeepMind researchers raise whopping $300M seed to automate science

Periodic Labs has launched from stealth with a substantial $300 million seed round, backed by a consortium of prominent investors including Andreessen Horowitz, Nvidia, and Jeff Bezos. Founded by former Google Brain and OpenAI leaders, the company aims to automate scientific discovery through 'AI scientists' operating autonomous labs, with an initial focus on inventing novel superconductors and other advanced materials. This initiative seeks to generate critical new physical world data to evolve AI models, positioning Periodic Labs as a significant, well-funded player in the burgeoning AI-driven materials science sector.

Analysis

Periodic Labs has emerged from stealth with a formidable $300 million seed round from a consortium of top-tier technology investors, including Andreessen Horowitz, DST, and Nvidia. The company's strategic focus is on automating scientific discovery through 'AI scientists' operating in autonomous laboratories, a capital-intensive model justified by its high-profile backing and the pedigree of its founders, who previously led key AI initiatives at Google Brain and OpenAI. The initial objective is to invent novel superconductors and other advanced materials, addressing a critical area of technological development. A core component of Periodic's strategy is to generate proprietary physical-world data from its experiments, aiming to overcome the data limitations of current AI models which have, in the company's view, 'exhausted' the internet. This positions the firm not just as a materials science innovator but also as a creator of a valuable, unique dataset for training next-generation AI. While Periodic Labs represents a significant concentration of capital and talent, it enters a competitive landscape that includes smaller startups and academic research consortiums pursuing similar goals in AI-driven chemistry and materials discovery.

Key Decisions for Investors

• Investors in the materials and specialty chemicals sectors should monitor Periodic Labs as a potential long-term disruptor, as its success could fundamentally alter the R&D and discovery timeline for new materials.
• The investment by Nvidia (NVDA) reinforces the thesis that AI applications in physical sciences represent a significant future demand driver for high-performance computing, validating continued exposure to key AI hardware enablers.
• This large seed round establishes a new valuation benchmark in the AI-for-science space; venture and growth equity investors should note the heightened institutional appetite and capital requirements for companies building proprietary physical data moats.
• For public market investors evaluating AI-centric companies, Periodic's strategy highlights the increasing importance of proprietary data generation as a key competitive differentiator beyond models trained on public data.