Andrea L. Thomaz

How do we make AI safe in healthcare without slowing down progress? In our latest Stanford Healthcare AI Podcast, we sat down with Shantanu Nundy, physician, technologist, and now FDA advisor on AI. In this episode we cover our typical updates on the state of AI and jump into some of the issues being wrestled with at the FDA on AI. A few highlights: 1️⃣ "What's the counterfactual?" Shantanu shared the five statistics he opens many FDA meetings with: 100 million Americans with no regular medical care, 75 million living in healthcare deserts, Medical error as the 3rd leading cause of death, 95% of rare diseases with no FDA-approved treatment, U.S. life expectancy 4 years behind other OECD countries. His point: “We’ve known for decades there’s a jumbo jet every day crashing due to medical errors.” Yet we hold AI to a higher bar than the unsafe status quo. A need for post-market monitoring 2️⃣ A growing emphasis on real-world evaluation Historically the FDA has been focused on pre-market testing and there is a clear need for real-world data. As models and deployments evolve, continuous learning and monitoring are essential. 3️⃣ The plumbing is missing Most health systems still can’t answer: Which AI tools are being used, where, and by whom? We need encounter-level tracking — timestamps, inputs, outputs, model versions — and system-wide visibility to make AI safety measurable. 4️⃣ Build on what already works Extend risk-based frameworks, device-style identifiers, and recall systems to AI. And if you’re working in this space, now’s the time to contribute to the FDA’s current RFI shaping what “safe and effective” AI looks like. The takeaway: AI is top of mind for the FDA and there is a clear need for new infrastructure. Thanks to co-host Matt Lungren MD MPH and the Stanford Center for Health Education | Professional Courses and Programs, and Stanford Online for helping to make this happen! Link to video in the comments.

150

12 Comments

When AI meets the physical world, science gets...automated? Periodic Labs just emerged from stealth with a remarkable $300 million seed round, backed by Andreessen Horowitz, NVIDIA, and Jeff Bezos. The company was founded by Ekin Dogus Cubuk, who led materials and chemistry teams at Google Brain and DeepMind, and Liam Fedus, former VP of Research at OpenAI and co-creator of ChatGPT. Their ambitious mission centers on automating scientific discovery through AI scientists operating in autonomous laboratories where robots conduct physical experiments, collect data, and iterate continuously. The startup's initial focus targets superconductor development, aiming to create materials that outperform existing options while potentially requiring less energy. What's particularly intriguing is their approach to the data problem facing AI development. As internet-trained models reach their limits, Periodic Labs plans to generate fresh physical world data through their autonomous labs, potentially fueling continued AI evolution while simultaneously discovering next-generation materials. Check out the article here: https://lnkd.in/eWC7wRBF If you like this content, consider subscribing to my weekly newsletter where I share 3 key events in the data and AI space, 2 BigQuery tips, 1 thing that piqued my curiosity - completely FREE. Follow the link below to sign up.☟ www.beardeddata.com/signup #PeriodicLabs #AIAutomation #MaterialsScience #ScientificDiscovery

1

Welcome to UT REAL Health AI! UT REAL Health AI brings together researchers, clinicians, technologists, and innovators across The University of Texas to advance the responsible use of artificial intelligence in healthcare. Our work focuses on bridging research and real world application by developing, evaluating, and implementing AI solutions that improve patient outcomes, support clinicians, and drive system-level impact. Learn more about our mission and ongoing initiatives at: https://lnkd.in/gPm69DSG Follow us for updates on research highlights, partnerships, and events in Health AI. #UTREAL #HealthcareInnovation #UTSystem #ArtificialIntelligence

47

1 Comment

Super thankful for my colleague and partner in crime Christian Gonzalez, PhD for pointing me in the direction of a topic to nerd myself over in the next few months - #LabAutomation 🧫🤖 Most deeptech conversations focus on scaling innovation. The underrated problem is *discovering* it faster. Cue in AI and Robotics. Some questions we are entertaining ourselves with at Planet A Ventures: 👉🏻 Is "Willingness to Pay" fundamentally different here compared to other sectors like manufacturing? My gut says yes - lab automation companies targeting emerging biotech or materials startups are dealing with clients who don't see R&D spending as overhead, but as the critical path to their next milestone. That makes for some pretty counterintuitive customer economics. 👉🏻 Who owns the data? If laboratories are seen as a space that generates data, and all the start-ups claim to have "proprietary data moat", did we really have a clear answer on who owns it? - the start-ups, the clients or the CROs? 👉🏻 Change Management - When robotic arms are dexterous enough to handle tiny samples and AI models can design complex experiments autonomously - what happens to the scientist at the bench? Automation efforts across every other industry have failed not because the technology didn't work, but because the humans around it weren't ready. Labs won't be different. Massive thank you for all the start-ups and experts that have shaped our thoughts the last month. Stay tune for more content on the topic! cc Alexander Hammer (Dunia), Philipp Heuermann (Ginkgo Datapoints), Robert Zechlin (UniteLabs), Lucas Mair (ProviGenAI), Karol Janik

59

5 Comments

At Texas A&M School of Engineering Medicine, Dr. Yanmin Gong is pioneering privacy-preserving AI systems that help hospitals learn from one another—without sharing sensitive patient data. By integrating federated, multimodal, and generative AI, she is shaping safer, more accessible technologies and preparing the next generation of physicianeers to lead in an AI-driven health care world. 🔗 https://bit.ly/4aScxkk #tamuenmed #AIinHealthcare #ResponsibleAI

29

2 Comments

The Texas Space Commission has approved a $14.1M SEARF award for Rice University to launch the Center for Space Technologies inside the Rice Space Institute. This is a serious and sophisticated funding decision that will turn deep-tech research into deployed capability, build an entirely new workforce sector, and create durable interfaces between labs, industry, and the state. For Houston, it is another step toward a fully integrated space economy. We already have human spaceflight, we already have the world’s densest clinical engine, and we already have a growing base of space companies that can build hardware at scale. What has been missing, historically, is the institutional flywheel that consistently converts scientific advantage into programs, talent, and operating infrastructure. Rice has been methodically building that connective tissue for years, and David Alexander deserves real credit for it. Institution-building is slow, compounding work, and this award is a signal that the state recognizes what Rice is assembling in Houston. If the Second Space Age is going to look like sustained operations rather than brief sorties, the center of gravity will shift toward the systems that keep humans healthy and productive when resupply, bandwidth, and medical expertise are constrained. That is where the next decade of breakthroughs will come from, and it is where Houston’s advantages start to stack in a way that is impossible to replicate elsewhere.

194

6 Comments