Close Menu

    Subscribe to Updates

    Get the latest creative news from FooBar about art, design and business.

    What's Hot

    Utah Launches First-Ever AI Prescription Pilot in the U.S., Sparking Debate on Safety and Innovation

    January 13, 2026

    EU Widens Tech Crackdown, Targeting Musk’s Grok and TikTok Over Alleged AI Law Violations

    January 13, 2026

    Malicious Chrome Extensions Compromise 900,000 Users’ AI Chats and Browsing Data

    January 12, 2026
    Facebook X (Twitter) Instagram
    • Tech
    • AI News
    Facebook X (Twitter) Instagram Pinterest VKontakte
    TallwireTallwire
    • Tech

      Malicious Chrome Extensions Compromise 900,000 Users’ AI Chats and Browsing Data

      January 12, 2026

      Wearable Health Tech Could Create Over 1 Million Tons of E-Waste by 2050

      January 12, 2026

      Viral Reddit Food Delivery Fraud Claim Debunked as AI Hoax

      January 12, 2026

      Activist Erases Three White Supremacist Websites onstage at German Cybersecurity Conference

      January 12, 2026

      AI Adoption Leaders Pull Ahead, Leaving Others Behind

      January 11, 2026
    • AI News
    TallwireTallwire
    Home»AI News»Ex-OpenAI/DeepMind Team Lands $300M Seed to Build AI Scientists
    AI News

    Ex-OpenAI/DeepMind Team Lands $300M Seed to Build AI Scientists

    Updated:December 25, 20253 Mins Read
    Facebook Twitter Pinterest LinkedIn Tumblr Email
    Ex-OpenAI/DeepMind Team Lands $300M Seed to Build AI Scientists
    Ex-OpenAI/DeepMind Team Lands $300M Seed to Build AI Scientists
    Share
    Facebook Twitter LinkedIn Pinterest Email

    A new startup called Periodic Labs—founded by former OpenAI and DeepMind researchers—has raised a massive $300 million seed round to build autonomous “AI scientists” that run experiments in physical labs, aiming to accelerate materials discovery and scientific innovation. The round was led by major names like Andreessen Horowitz, DST, and Nvidia, with backing from Jeff Bezos, Eric Schmidt, and others. The company intends to go beyond training models on historical scientific literature, instead combining AI with robotic laboratories to design, test, and iterate novel materials (e.g. next-generation superconductors) while generating fresh experimental data. Meanwhile, institutions such as MIT have unveiled AI systems like CRESt that already perform experiments and discover new materials. Across labs and startups, the push to automate science — from chemistry to materials engineering — is heating up.

    Sources: MIT.edu, AI Insider

    Key Takeaways

    – The $300M seed round for Periodic Labs marks one of the largest early bets yet on automating lab science rather than only training on existing datasets.

    – By pairing AI decision-making with physical experimentation, the startup aims to generate proprietary data that models don’t yet have—potentially outpacing what purely computational or literature-driven systems can produce.

    – Parallel academic efforts (such as MIT’s CRESt) and other AI-science automation systems suggest that the convergence of robotics, modeling, and data generation is becoming a mainstream trajectory in scientific R&D.

    In-Depth

    This move by Periodic Labs is part of what’s becoming a broader shift in how science can be done: not simply by human-led theory, experiment, and iteration, but via AI agents backed by robotic labs that can close the loop themselves. Until now, most AI in science has involved training on published data, literature, and simulation outputs—but that approach inherently limits what the model can discover. Periodic intends to change that by letting its AI control real laboratory hardware, running syntheses, measurements, and modifications autonomously. Their first target: next-generation superconductors, materials in which even minor improvements can have big payoff in energy, computing, and infrastructure.

    The founders bring serious credentials: Ekin Dogus Çubuk was with DeepMind and Google Brain, leading materials and chemistry work, while Liam Fedus was a research VP at OpenAI and one of the minds behind large models. Their thesis is that scientific AI has hit diminishing returns when limited by past data, and the future lies in systems that can generate new data. They plan to have their autonomous labs iterate experiments, learn from outcomes, and refine hypotheses—a closed feedback loop between AI design and physical realization.

    This is not just speculative. On the academic side, MIT researchers recently unveiled “CRESt,” an AI platform that can ingest multiple modalities of scientific information and propose experiments, then run them to discover new materials. That model echoes the same principle: an AI that doesn’t just analyze, but executes and learns. At the same time, foundational work in “self-driving” chemistry and materials science is advancing, exploring how robotics and AI can accelerate discovery. The scientific community is grappling with challenges: how to ensure safety, reproducibility, interpretability, and integration into existing research pipelines. But investors seem convinced: a large infusion of capital into Periodic Labs signals belief that autonomous science could redefine how breakthroughs are made.

    If they succeed, the implications are huge. Faster materials discovery might spur leaps in energy efficiency, computing power, battery design, sensors, and even pharmaceuticals. The era when machines were assistants to scientists might soon shift to machines as active scientific agents themselves.

    Share. Facebook Twitter Pinterest LinkedIn Tumblr Email
    Previous ArticleEx-NotebookLM Engineers Launch “Huxe” to Turn News & Research into AI-Generated Podcasts
    Next Article Executive Exodus Deepens at Hyundai’s Air-Taxi Venture Amid Program Pause

    Related Posts

    EU Widens Tech Crackdown, Targeting Musk’s Grok and TikTok Over Alleged AI Law Violations

    January 13, 2026

    Utah Launches First-Ever AI Prescription Pilot in the U.S., Sparking Debate on Safety and Innovation

    January 13, 2026

    Microsoft Warns of a Surge in Phishing Attacks Exploiting Misconfigured Email Systems

    January 12, 2026

    Malicious Chrome Extensions Compromise 900,000 Users’ AI Chats and Browsing Data

    January 12, 2026
    Add A Comment
    Leave A Reply Cancel Reply

    Editors Picks

    Malicious Chrome Extensions Compromise 900,000 Users’ AI Chats and Browsing Data

    January 12, 2026

    Wearable Health Tech Could Create Over 1 Million Tons of E-Waste by 2050

    January 12, 2026

    Viral Reddit Food Delivery Fraud Claim Debunked as AI Hoax

    January 12, 2026

    Activist Erases Three White Supremacist Websites onstage at German Cybersecurity Conference

    January 12, 2026
    Top Reviews
    Tallwire
    Facebook X (Twitter) Instagram Pinterest YouTube
    • Tech
    • AI News
    © 2026 Tallwire. Optimized by ARMOUR Digital Marketing Agency.

    Type above and press Enter to search. Press Esc to cancel.