What will the next 3–5 years of AI in health and life sciences actually look like? At World Summit AI Amsterdam, Daphne Koller, Founder & CEO of insitro, Co-founder of Coursera and former Stanford Professor, will lead an exclusive 60-minute executive masterclass on AI for Health and Humanity. This isn't a traditional keynote. It's an interactive conversation with one of the pioneers of modern AI, built around the questions senior leaders are actually wrestling with as AI reshapes human biology, healthcare and drug discovery. The conversation will explore: → What comes beyond foundation models → Where AI is creating real value in biology and healthcare, and where the hype breaks down → The next wave of AI companies: what to build, what to buy and how to tell the difference → What health and life sciences executives most often get wrong about AI → How regulation, capital and data will influence what wins → A candid look at the next 3–5 years, including which bets could pay off and which fashionable ideas may not age well One hour. A limited room. Direct access to one of the people who has helped define modern AI. 📅 Thursday 8 October 2026 ⏰ 2:30–3:30 pm 📍 Taets Art & Event Park 🎟️ €199 | Registration required ⚠️ Attendees must hold a valid World Summit AI Expo & Content ticket. Places are limited and booking closes 25 September. Register now: hubs.li/Q04x6yjH0 #WorldAIWeek #WorldSummitAI #DaphneKoller #AIForHealth #HealthcareAI #DrugDiscovery

Biopharma is rapidly embracing AI-native drug discovery, which is opening up some interesting new avenues for collaboration. We @a16z are co-hosting an #SFTechWeek breakfast with @EliLillyandCo to dig into what this looks like in practice. We’ll hear from leaders at Lilly TuneLab, @insitro, @BigHatBio, Flex Therapeutics, and a16z about how federated learning, biology foundation models, generative design, and closed-loop experimentation can help our industry make better predictions and accelerate drug discovery. Come hang out with us! @Techweek_ @VanessaBraunst partiful.com/e/ccQ8p5st5UNpG…

H @DaphneKoller ιδρύτρια της @coursera και σήμερα της @insitro εξηγεί στο WIRED Greece γιατί η εμπλοκή του AI στην ανακάλυψη φαρμάκων αφορά τον σχεδιασμό καλύτερων μορίων αλλά και την αποκάλυψη των βιολογικών μηχανισμών που ακόμη δεν κατανοούμε. wired.com.gr/article/i-daphn…

insitro is thrilled to announce the appointment of Hideo Makimura, M.D., Ph.D., as Chief Medical Officer, a pivotal step as we advance our therapeutic programs toward the clinic. "Hideo has spent his career using human genetics and biomarkers to guide and accelerate clinical programs from initial patient studies through late-stage development," said Daphne Koller, founder and CEO of insitro.” “As insitro heads into the clinic, I could not imagine a better partner to help us translate the insights generated by our platform into meaningful medicines for patients." Learn more: insitro.com/news/insitro-app…

AI biotech Insitro’s Daphne Koller on how to speed up clinical trials By @brittanytrang statnews.com/2026/08/26/biot… via @statnews

Dear followers, this is a fascinating, thought-provoking article on AI and drug discovery. The facts are very interesting. I suspect the pattern that Daphne Koller documents may apply more broadly than just to drug discovery. deepphenotype.substack.com/p…

Excellent piece on the deep challenges of drug discovery and the ways that AI can and cannot help, from Stanford Comp Sci professor Daphne Koller, founder of Coursera and now Insitro, an ML-enabled drug discovery company. deepphenotype.substack.com/p…

Mapping Causal Human Biology - @DaphneKoller @insitro deepphenotype.substack.com/p…

The second issue of Deep Phenotype, insitro’s Substack publication on scaled biology and deep causality, is now live. In “Mapping Causal Human Biology: From Outcome to Mechanism,” insitro CEO and founder @DaphneKoller examines the problem at the center of drug discovery – moving from a desired clinical outcome to a mechanism precise enough to drug. Roughly nine in ten drugs that enter the clinic fail, most because the underlying biology was wrong. Today’s AI can detect associations at an extraordinary scale. Drug discovery, however, demands something harder: causal knowledge of what will happen when we intervene in a human system. Human genetics offers rare evidence from experiments nature has already run. Targets with genetic support are two to four times more likely to succeed in the clinic, yet only 3.6% have been pursued for any supported indication. Read Daphne’s new article to learn how insitro approaches this opportunity – and why getting the biology right matters far beyond the probability of success. “Every trial asks patients to accept real risk on an unproven hypothesis. When the biology was wrong from the start, that risk was taken for a drug that was never going to help them. Better target selection honors what patients put on the line.” ╰┈➤ Read part two of Daphne’s manifesto on Substack, “Mapping Causal Human Biology: From Outcome to Mechanism” on Deep Phenotype: deepphenotype.substack.com #DrugDiscovery #AI #Biology #CausalBiology

A masterpiece from @DaphneKoller — honestly a must-read for anyone who claims to be deeply interested in the role that AI will play in drug discovery and development @insitro is one of the most AI-pilled biotech companies in the world, using agents and AI tools day in and day out across their organization for myriad tasks and workflows But they are also one of the few AI x biotech companies who have committed to actually developing novel medicines, against novel targets. The team has seen first-hand how hard (and often incompressible) every step of this process is. Daphne shares thoughtful reasoning about where magic is possible, and where it probably isn’t (at least not yet). a16z.news/p/drug-discovery-h…

“AI will mostly help us generate failures faster.” A provocative take from Daphne Koller, Founder & CEO of insitro and World Summit AI 2026 speaker, on what AI can — and can’t — solve in drug discovery. Read Drug Discovery Has No Magic Wands ↓ hubs.li/Q04sV8yx0

insitro founder and CEO @DaphneKoller recently stopped by The Information Bottleneck podcast to chat with hosts @ziv_ravid and @Hellisotherpe10 about causal biology and why AI in drug discovery is a fundamentally different problem than AI in language or vision. The conversation covers a lot of ground: → The bitter lesson worked for language and vision because scale and general methods met an internet's worth of data. Biology has no equivalent corpus — cells grow at the speed cells grow. AI cannot manufacture the missing data. → A drug isn't a pattern in data you already have, it's a prediction about an intervention nobody has run yet. This is a significant reason why 90% of drugs that reach the clinic fail, and only 22% of diseases have any approved treatment at all. → Plus: what it would take to build that missing corpus, and what real foundation models for biology would require. 🎧 Listen to the podcast at theinformationbottleneck.com… 📖 Daphne's article “Drug Discovery Has No Magic Wands," on insitro’s Substack, Deep Phenotype, covers similar territory in greater depth — read and subscribe: deepphenotype.substack.com/ #AI #DrugDiscovery #Causality #Biology

Super clearly articulated call to action by @DaphneKoller @insitro “AI is undoubtedly extraordinary, but aimed at a biology we have only begun to measure and barely understand, it will mostly help us generate failures faster.” “The real bottleneck in making a novel medicine is disease understanding: identifying a biological mechanism whose modification actually changes the course of disease in patients.” “The diseases where we have made the least progress tend to be precisely those that are most human-specific, and therefore those for which the data is most expensive to collect, least available, and most fraught with ethical constraints.” open.substack.com/pub/a16z/p…

A machine learning pioneer was asked why AI has not cracked drug discovery. She did not blame the models. Daphne Koller wrote the textbook a generation of ML researchers learned from and now runs insitro. She says 90% of drugs that enter the clinic fail, and the vast majority fail for a reason that has nothing to do with chemistry. The molecule does exactly what it was designed to do. The mechanism it hits turns out to do nothing for the disease. Only 22% of diseases have any FDA approved drug at all. She calls that 22% the upper bound of what we understand, not the floor. Biology's bottleneck is knowing what to aim the model at.

Only about a quarter of human diseases have an approved therapy. By some estimates it's a few percent. Most of those treatments slow a disease rather than stop it. Closing that gap is what the AI-cures-everything story promises. Build a system smart enough and the cures hidden in what we already know will fall out. I have worked at the intersection of machine learning and biology for three decades, and I believe #AI will eventually transform human health. The capabilities arriving now are extraordinary. But the promise rests on an assumption that is simply false: that we already understand human biology well enough for a clever enough reasoner to find the answers in it. We don't. More than 90% of drugs entering clinical trials fail, a number that has barely moved in decades. In the large majority of those failures the molecule was engineered just fine. The mechanism it targeted was wrong. We are doing a pretty good job at manufacturing keys, but they are generally for the wrong locks. And because nobody wants to fail in the clinic, the industry has retreated to the locks it already trusts: 38 targets now have more than 50 programs against each of them, while the number of novel targets advanced per year fell from roughly 100 in 2015 to about 30 in 2024. AI will not reason its way past this. Biology wasn't engineered. It is the product of billions of years of messy, stochastic evolution, and the variation that produced is too vast and too idiosyncratic to work out in the abstract. You have to measure it. Aimed at a biology this thinly sampled, AI will mostly help us generate failures faster. I founded @insitro because getting to the right locks requires a different kind of system. We generate multimodal human and cellular data at scale, use machine learning to find causal drivers of disease, and test those hypotheses experimentally. Virtual Human™ is built for causal discovery; TherML™ turns what it finds into the right therapeutic intervention. It is working: first-in-class programs internally and with partners, three #ALS targets that Virtual Human™ identified and @bmsnews nominated, and additional collaborations with @EliLillyandCo and @GileadSciences . Today we are launching Deep Phenotype: Scaled Biology, Deep Causality. Issue one is "Drug Discovery Has No Magic Wands," the first half of a two-part essay on the magical thinking currently running through our field and what I think it will actually take. After that you will hear from insitro's own scientists and engineers, people who work across computation and experiment because the problem requires both. Getting this right is hard, and we do not have all of it worked out. I hope you will follow along and think it through with us. DeepPhenotype.Substack.com

The data we need to understand human biology is about 1000x the quantity we've collected Full piece from @insitro founder and CEO @DaphneKoller: a16z.news/p/drug-discovery-h…

Exciting day at insitro with the launch of Deep Phenotype, our new Substack publication covering scaled biology and deep causality. Over the coming months, we'll separate truth from hype as we explore causal human biology, physical AI, and the work of turning discovery into medicine. Issue No. 1 features part one of a two-part manifesto written by insitro founder and CEO @DaphneKoller. Follow the link to read Drug Discovery Has No Magic Wands, and subscribe to Deep Phenotype to get part two directly in your inbox. - Read and follow: deepphenotype.substack.com/

This summer, insitro has been traveling the globe to showcase our physical AI platforms for causal human biology and drug discovery. → #ICML2026 Seoul: Emily Fox on causal validity in ML-driven drug discovery → #SBDD Summit Boston: Nikhil Jain on construct strategy, stability, and biology for modern structure-based drug design. → AIMLR 2026 Rome: Aj Kaykas on ML-enabled ALS target discovery → Stanford Cardiovascular Research Symposium Palo Alto: Daphne Koller on Causal AI for Human Health: Engineering the Discovery of Better Medicines Up next: → 18th Annual Bioprocessing Summit, Aug. 10: Jennitte L. Stevens

I agree completely, and I would push it further: every company is now a data company, deliberately or not. Some are just more explicit about it than others. In science we make it literal: we don't generate that knowledge as exhaust from using a model, we generate it directly — causal insight into human biology, measured by intervening, not inferred from anything a model was trained on. And that over-layer — deliberate data generation, proprietary workflows — is how you turn a generic intelligence model into one where causality is a first-class citizen. That over-layer is an organization's intelligence, and it goes far beyond the frontier model. It lives in the proprietary data, specialized tools, the multi-step workflows, the evals and data, and the people who wire them together — and that composite is the proprietary core. The model is the swappable layer; the loop that compounds the rest is the "veteran" capability that stays when the generalist model changes, and gets smarter every turn of the crank. Which is exactly @satyanadella's point: in using intelligence, you create intelligence — and what you create should be yours.