Markus J. Buehler

The next frontier in protein design will not be defined by structure alone, but by the capacity to engineer motion as a first-class principle of function. This is because dynamics is where the real biology lives. Foundational work by Karplus, Levitt & Warshel made clear that chemistry cannot be understood without motion, mechanism, and scale. Gō, Brooks & others showed that proteins possess characteristic collective motions - low-frequency normal modes that capture how whole molecules bend, breathe, and fluctuate. Frauenfelder then sharpened the picture further: proteins are not static objects occupying a single minimum, but dynamic ensembles traversing rugged energy landscapes. And yet the modern AI revolution in protein science has been, above all, a revolution in structure. In our new paper in Matter, @_Bo_Ni and I ask a different question: not what structure will this sequence adopt? but what sequence will realize a prescribed pattern of motion? VibeGen inverts the conventional design paradigm. Rather than treating dynamics as a consequence to be analyzed after the fact, it makes dynamics the design objective from the outset. Using a language diffusion model with two cooperating agents - a designer that proposes sequences and a predictor that critiques them against the target motion profile - the system converges on de novo proteins with tailored vibrational behavior. One of the most intriguing results is a form of functional degeneracy - distinct sequences and distinct folds can satisfy the same target dynamical specification. For a given functional pattern of motion, evolution may have sampled only a small region of the physically realizable design space. The space of viable molecular mechanics may be far larger than the repertoire biology happened to discover. We have made "vibe" into a cultural metaphor - something intuitive, affective, subjective. But at the molecular scale, vibe is not metaphor: It is physics. For a protein, the vibe is the pattern of motion itself; the fluctuations, resonances, and collective displacements that determine what the molecule can do.

A new model for scientific discovery is taking shape. The Genesis Mission combines AI, world-leading computing, and the expertise of our National Labs to help accelerate breakthroughs in materials, chemistry, quantum, and our understanding of the universe. energy.gov/undersecretary…

Fun panel coming up.

@james_y_zou @mertyuksekgonul Congratulations!

TTT-Discover is accepted by #icml2026 as a spotlight!🚀 Great job @mertyuksekgonul and team!

What happens when you combine AI supercomputers, quantum computing, and robotic laboratories? You get the Genesis Mission. At the #AIExpoDC, Under Secretary Darío Gil (@ScienceUnderSec) explained how this new national effort will double the impact of America’s $1 Trillion R&D engine within a decade. It's time to take AI beyond language and into the realms of physics, chemistry, and biology. 🧬💻 Watch his full breakdown here: youtu.be/I-gCjfK4tng #DOE #DaríoGil #TechInnovation #DepartmentOfEnergy #QuantumComputing #ArtificialIntelligence

This is protected via gates: E.g., minimum description length (MDL) on the Builder (every new object/morphism has to pay for itself in description length!). Adversarial pressure from the Breaker - here, a new structure has to survive falsification against simulation or measurement. And compositional consistency, which means that the enlargement has to extend the old category, not overwrite it. So, drift is permitted but expensive.

@ProfBuehlerMIT @BrownUniversity @ICERM If fracture introduces new admissible objects and morphisms, what constrains category expansion from becoming unconstrained semantic drift?

Yesterday at @BrownUniversity @ICERM's workshop on “Agentic Scientific Computing and Scientific Machine Learning” I spoke about “Adaptive Swarms Across Scales”, making the case for scientific AI as systems that can create representations, stress them, fracture them, and enlarge the category in which future representations live. The category here is a composable and breakable working universe of science: data, hypotheses, simulations, measurements, tools, failures, figures, papers, provenance, and the transformations that connect them. Discovery happens when those transformations become executable, inspectable, composable, and capable of changing the world model they operate within. Atomistic modeling gives one category - states, forces, trajectories, observables, boundary conditions, conservation laws. Neural surrogates learn fast morphisms inside or between such categories. But discovery is higher-order: it changes which objects and morphisms are available in the first place: what variables exist, what operations are allowed, what evidence counts, what scale is active, what invariant is being preserved, and what kind of explanation the system is even capable of forming. This is scientific method as adaptive architecture: compression, stress, fracture, recomposition. Fracture matters here because it makes the logic physical: a non-commuting diagram realized in matter. The imposed load, material hierarchy, defect field, and assumed continuum description no longer map cleanly into the observed outcome. The crack is the obstruction and it identifies where the old morphism failed and where a new representation must be introduced. The physical crack and the categorical obstruction are the same event viewed in different substrates. ScienceClaw × Infinite is a machine for constructing and transforming a category of scientific artifacts. Each artifact is typed. Each operation has lineage. Each failed branch remains in the category as reusable structure. The “paper” is no longer the terminal object of science; it is one projection of a larger compositional trace, and it can be generated at any time for consumption by a human or an AI. With that the unit of scientific labor is changing. For most of the twentieth century the unit was the result (a measurement, a theorem, a synthesized molecule). It is now becoming the algorithm that produces results, and after that, the substrate of discovery itself. The static PDF is the wrong terminal object for this regime, and the role of the scientist with it. We now design algorithms that build algorithms, and eventually substrates in which such algorithms compose themselves. At that point, the scientist is no longer outside the discovery system. The scientist becomes one of the representations the system can transform. In that sense, the systems will eventually do science to us, and that is the structural consequence of the principle they are built on.

Great thread on how the scientific process could evolve with agentic learning.

We've anticipated it, but it's now here: METR shows that frontier LLM task horizons are doubling every ~105 days and @AnthropicAI's Mythos model breaks the benchmark. For AI-for-science this makes agents and swarms exponentially more powerful - a faster substrate means each member of a swarm carries a deeper hypothesis and execution capability. ⤵️

We’ve expected this but still incredibly exciting to see ⤵️

Read more in our technical paper: arxiv.org/abs/2605.06651 See the Frontier Tier 4 problem evaluation at: epoch.ai/frontiermath/t…

Very important read. "if AI mathematics continues to progress at anything like its current rate -- which is what I expect to happen -- then we will face a crisis very soon"

Creativity is novelty lived in time, by a system that cannot fully precompute its own becoming, and is changed by what it puts forth. In other words: creativity requires a system that unfolds in time, cannot foresee its own outputs, and is transformed by what it creates. Honored that John Werner @Link_Ventures featured my work in @Forbes. forbes.com/sites/johnwern…

👉🏼 highly suggested short course on #AI, #manufacturing & #materials by @ProfBuehlerMIT !!!

MIT Generative Multiscale Materials Design: Physics, AI, Manufacturing Short Course June 1-4, 2026, MIT Campus OR Live Online, Cambridge, MA TL;DR: Deeply embedded at MIT, this course offers a high-impact week of technical lectures, hands-on labs, and interactive clinics focused on the future of agentic materials discovery. You will master cutting-edge workflows using physics-based generative AI, multi-agent systems, molecular modeling, and precision manufacturing - culminating in an official MIT certificate. This short course immerses you in a dynamic environment featuring technical lectures, group design studios, interactive labs, and participant talks. It is a unique opportunity to not only learn but to network with global peers and MIT researchers. You will move beyond simple prediction to master autonomous AI workflows. Through hands-on clinics, you will build multi-agent systems that reason, plan, and invent next-generation smart materials. The curriculum is designed to accelerate your ability to leverage the most in-demand areas of materials engineering: 1️⃣ Generative Multiscale Modeling: Bridge atomic-level insights to macroscopic performance using "Scientist" and "Critic" agents to hypothesize and validate concepts. 2️⃣ AI for Science: Master Large Reasoning Models, Diffusion Models, and Graph Neural Networks (GNNs) for autonomous discovery. 3️⃣ Bio-Inspired Design: Utilize category theory and bio-knowledge graphs to transfer nature’s design principles into synthetic materials. 4️⃣ Autonomous Manufacturing: Execute "bit-to-atom" workflows, using AI to drive multi-material 3D printing and physical validation. 5️⃣ Nanotechnology: Engineer function and performance at the smallest scales using bottom-up construction. Instructor: Markus J. Buehler, McAfee Professor of Engineering, MIT

The future of AI and energy is happening at @SCSP_AI's AI+ Expo! Discover the Genesis Mission, America's flagship science and tech initiative uniting all 17 of @ENERGY's National Labs to accelerate AI-driven scientific discovery in DC, May 7-9. Speaker & demo schedule below!

We managed to line up another fantastic speaker from out of town this month thanks to #PEGSBoston! Come see Ariel Tennenhouse present on Wednesday, May 13th 2026 at 7pm EDT in Room 181, Building 68, MIT "Computational design of antibody repertoires" bpdmc.org

@FutureJurvetson @jendionne @FutureSaenko Congrats @jendionne & team!

🐠 Everything we know about biology has been built on an incomplete picture. DNA tells us what a cell might do. Proteins tell us what it’s actually doing. Pumpkinseed announced their $20M Series A today (led by Future Ventures and NfX) to build the platform that reads proteins directly—for the first time. Proteomics has always faced a fundamental constraint: you can only measure what you already know to look for. The current workhorse, mass spectrometry, requires matching protein fragments against reference databases. If a protein isn't in the database, or doesn't ionize reliably, it's invisible. Other approaches rely on fluorescent labels or antibody-based affinity methods, which introduce their own biases and blind spots. The result is a field that has spent decades generating an increasingly detailed map of a small, well-lit corner of the proteome, while biology’s most important data layer remains hidden. This isn't a sensitivity problem. It's a category problem. Existing tools were never designed to read proteins directly de novo. They were designed to find what researchers already suspected was there. Pumpkinseed is built to find everything else. And proteomics is harder than most people outside the field appreciate. When we account for post-translational modifications, non-canonical amino acids, and glycan decorations, there are roughly a thousand distinct chemical monomers in the proteomic alphabet, compared to the four bases of DNA. deSIPHR (de novo Sequencing and Identification of Proteins with High-throughput Raman spectroscopy) is Pumpkinseed's proprietary nanophotonic chip platform, fabricated with semiconducting manufacturing. With over 100 million sensors per square centimeter, it reads proteins, known or unknown, letter by letter — amino acid by amino acid — without a reference catalog of proteins, and at high-throughput. The result is direct, high-resolution proteomic data, including post-translational modifications, non-canonical amino acids, and single-cell detail, that mass spectrometry-based approaches cannot match. What is Raman spectroscopy? Rather than tagging or fragmenting proteins, Raman spectroscopy reads the molecular vibrations of individual molecules. Each amino acid vibrates at a characteristic frequency, producing a unique physical signature that deSIPHR detects directly. This is physics reading biology in the most literal sense. With conventional Raman spectroscopy, only about one in ten million photons interacts with a molecule usefully, far too weak for single-molecule work. Pumpkinseed's answer is a silicon photonic chip patterned with a billion sensors per wafer. Those sensors concentrate light into volumes smaller than a single protein, amplifying Raman scattering efficiency by over 10 million-fold. And their future ventures? “The longer-term ambition is the virtual cell, a computational model that simulates not just how proteins fold but how they interact, respond to drugs, and behave under perturbation inside a living system. AlphaFold demonstrated what structural AI can do once a sequence is known. The gap that cannot be closed is determining the sequence itself from biological samples, particularly for proteins carrying modifications absent from existing databases. Pumpkinseed is designed to supply that input layer. "If the Human Genome Project was the data infrastructure that enabled genomic medicine, we believe the high-resolution proteomic dataset Pumpkinseed is building could be the analogous foundation for AI-driven biological discovery," co-founder Dr. Jen Dionne says. "In our vision, the molecular signatures driving disease, aging, and ecosystem health become fully legible. Medicine shifts from reactive to proactive. Optimal healthspan moves from aspiration to achievable reality." —synbiobeta.com/read/pumpkinse… • The biology mining company: Pumpkinseed.Bio • Today’s News: pumpkinseed.bio/news/pumpkinse…

Congratulations @stephenbalaban you have built an amazing company!

Introducing Altara: the scientific intelligence platform for the physical world. Today @evatuecke and I are excited to announce our $7M seed led by @GreylockVC, joined by @Neo, @BoxGroup, @Liquid2V, and angel investors including @JeffDean and leadership from OpenAI & AMD. We’re already working with early customers in semiconductors, batteries, and advanced materials. More below.