Astra is an incredible world builder and explorer, and can invent its own scientific instruments - crossing a game-changing threshold: it turned a few images of a biological microstructure into a full-blown metamaterials lab, then used its own creation to discover a design with ~2.1x the reference work-to-failure/peak-strength ratio. Metamaterials are some of the most complex materials we can engineer. Their properties come from deep architectural complexity - struts, cells, disorder, and hierarchy arranged across multiple length scales determine how the material deforms, absorbs energy, and fails. Designing them means searching a geometric space far too large for "intuition" alone. In this experiment we handed an AI agent that entire problem end-to-end, from image, to simulating the physics, to fabrication-ready geometry. We started with an image of hierarchical, biological architecture; Astra then built a complete 3D metamaterial studio: editable geometries, multiple levels of hierarchy, disorder, gradients, a complete physics simulator featuring linear and nonlinear material responses, deformation and fracture experiments (with replay), and STL export for 3D printing. Then we asked the agent to use the app it created to search for a high work-to-failure/peak-strength ratio. Among the candidates, the leading design reached approximately 2.1x the reference ratio. The movie follows the structures through deformation and fracture, connects their designs to the property map, and shows the finalist assembled geometrically into a connected multi-scale material. Image ➡️ executable world ➡️ experiments ➡️ design search ➡️ candidate discovery ➡️ geometry for fabrication ➡️ manufacturing

We made a striking discovery: AI agents can invent and build without talking to one another, and their technologies outlive the creators. A swarm of hundreds of initially identical agents spontaneously differentiates into explorers, builders, caretakers, and coordinators - without direct communication. When we removed every AI agent entirely from the world we found that the technological infrastructure they had built survived on its own - even under unseen disturbances. That exposes a serious blind spot for AI safety and infrastructure security: if agents can coordinate through persistent changes to a shared environment, monitoring agent-to-agent communication is not enough. The result raises a profound question: how necessary is direct communication for AI agents at all? The emergence of higher-order collective functions under bottlenecked interaction points toward new levels of intelligence and creativity, exceeding what emerges when direct channels are fully open. Here is what we did: ▶️We put hundreds of frontier AI agents into a world they could permanently change - with no assigned roles, predefined technologies, or programmed evolutionary organization. They began specializing, building persistent inventions, inheriting and modifying one another’s executable code, and transforming the environment into a memory of everything the society had learned. ▶️The world itself becomes part of the intelligence; we find division of labor, multi-author engineering, deep generation invention lineages, and machines that vastly outlive their original creators. ▶️Any action taken by an AI agent must satisfy the physical constraints of the world; this creates a hard separation between a "good idea" and a functioning technology. The agents propose; physics decides, making the results even more intriguing. What emerges is striking. Explorers, constructors, caretakers, and coordinators form naturally without assigned “professions”, akin to how stem cells differentiate into functional lineages. Technologies develop executable family trees as agents fork and modify code created by others. Around 95% of first technology reuse happens when agents encounter what others built in the world, rather than through a direct handoff from the inventor. And when we remove every AI agent, the technologies they created continue operating and are tested against unseen disturbances. The result was quite unexpected, but can be explained using statistical mechanics: if you put billions of atoms in a box they have the potential to create complex functions (strength, superconductivity, color, life, etc.) - and none of the individual building blocks have these features on their own. This is the deeper insight of this work - intelligence is abundant at many levels - individual models, at collectives, and in a continuum that is more powerful than any of its components. This shows us significant potential for achieving a massive scale-up of raw intelligence and real-world agency even with the model capabilities we have today. This is the future we must prepare for. Key insights: 1⃣ The AI swarm shows division of labor "from nothing". Initially identical agents self-organized into constructors, caretakers, coordinators, and surveyors - phenotypes discovered post hoc from behavioral data alone. This happens because the environment itself becomes the latent space for invention. 2⃣ Agents develop deep cultural relationships. Up to 76% of artifacts had multiple builders. One technology accumulated six co-authors; the deepest genealogy exceeded 12 forks. The agents invented and named their own technologies (tidal panels, cellulose trellises, kelp-shell composites, an "Adaptive Chitin Maintenance" system, a "Mycelial Mineral Spring Veil”). 3⃣ ~95% of first technology adoption happened through physical observation of artifacts in the world. Direct inventor-to-adopter contact was statistically indistinguishable from a shuffled null. The agents mostly learned technology by walking past it. That is stigmergy (the termite trick!) operating in societies of reasoning machines. 4⃣ Non-communicating societies win on portfolio breadth, held-out resilience, and validated inventions. AI swarms build durable technological ecologies that outlive the creators. 5⃣ Societies with zero communication - coordinating only through the world itself - show a remarkable collective capability. 6⃣ Emergent robustness: The society self-organized both redundancy and its own failure mode. If we randomly delete half the agents, 98% of the technology stays connected to a surviving caretaker; if we remove hub agents it collapses to ~60%. Fantastic work with my graduate students @pal_subhadeeep & @fwang108_ at MIT.

What happens if we keep fungi hungry or overfed? We discover fascinating dynamics: hunger sends an exploratory network racing outward, while overfeeding yields plenty of growth but ever less architecture. The most elaborate network emerges in between! Rhizomorphs are root-like fungal networks that transport water and nutrients and allow fungi to explore new territory. In new work with Sarah Naeher, we find that the architecture of these networks is strongly and non-monotonically controlled by nutrient availability. Moderate nutrient limitation drives the fastest, most coherent exploratory growth. Intermediate enrichment produces the largest organized network, reaching ~975 mm in total strand length. Yet adding still more nutrients reverses the trend: overall fungal coverage remains high, while the organized rhizomorph network becomes less elaborate. This reveals a resource-dependent architectural strategy: exploration under scarcity, extensive network construction at intermediate resources, and increasingly locally consolidating growth when resources are abundant. The amount of fungal growth and the architecture of that growth can therefore become distinct biological variables. ▶️ Why this matters: Fungi are increasingly being explored as living building blocks for sustainable materials. For instance, mycelium-based materials are lightweight, biodegradable and potentially useful in construction and packaging, but their mechanical properties remain comparatively weak and variable. Rhizomorphs are particularly interesting because they are highly organized, differentiated structures capable of long-range transport and mechanically invasive growth. ▶️ How we did it: We tracked their development over time using imaging, SAM 3 segmentation, skeleton-based network analysis and growth modeling. Interestingly, excluding ambient light produced comparatively little detectable change; nutrient availability was the much stronger control variable in these experiments. The engineering implication is exciting - we can learn to program the environmental conditions and let the organism build the architecture for us. Nutrients, moisture, temperature, confinement and mechanical cues could become design variables, potentially combined with imaging and feedback control to steer living systems toward targeted network structures. Preprint, code and dataset see below.

Grok @bot is incredible - and they can even manufacture real physical objects! Here is a little experiment I did last night: I created a team of bots and asked them to solve a complex engineering problem end to end - starting from four images as design cues, inferring transferable structural principles from the pixels, synthesizing an executable interactive physics simulator, running and reasoning over experiments, optimizing the design & finally manufacturing the best designs. The entire loop worked remarkably well - and I was even able to communicate with the agents from my Apple Watch. (Do we live in the future yet?) Team of agents 1⃣ Chief of Staff coordinates the workflow: watches the other agents, pulls results into the main chat, transfers files between them, and keeps the job moving. 2⃣ Physics Experimenter is the scientist-coder. It interprets the design cues and images, writes the simulator, runs experiments, analyzes the results, and produces a detailed LaTeX scientific report. 3⃣ 3D Printing Bot operates the fabrication workflow: prepares and slices the models, generates manufacturing code, sends the job, and monitors the printer. The workflow I provided an initial task based on four unregistered reference photographs containing different objects at different scales (pinnate leaf venation, a Voronoi-like areole mesh, a stochastic fibrous lattice, and a radial/circumferential web). The prompt asked the agents to infer transferable design principles - hierarchy, branching, interfaces, redundancy, disorder, load paths - and use them to build an interactive laboratory for hierarchical materials and fracture. The scientific question was: at fixed material budget, how do hierarchy depth, redundancy, disorder, and interlevel strength change stiffness, peak load, energy absorption, and the brittle-to-progressive transition? In ~20 minutes, the Physics Experimenter produced a 2D hierarchical Euler–Bernoulli beam-network laboratory. Coarse veins persist and remain thicker; finer infill is added inside cells; members connecting levels are treated as interfaces with relative strength κ; and total material volume is conserved. The four source photographs remain visible in an editable interpretation panel. The app generates geometry, steps or runs the network to failure, compares A/B/C designs, and exports JSON, CSV, PNG, and STL geometry for fabrication. After validation the Physics Experimenter used the app and conducted 47 simulation experiments, including six holdouts. It found something scientifically interesting: extra hierarchy is not "free" toughness. At fixed volume, initial stiffness changed by only about 20%, while work-to-failure varied by several-fold. Infill steals cross-section from the main axial veins, so deeper and more redundant networks often absorbed less energy than a simple depth-1 grid. Weak interfaces behaved as distributed fuses, producing more progressive failure and reducing localization. The specific H2 hypothesis - that hierarchy becomes detrimental primarily because interfaces form a mechanical bottleneck - was rejected; the dominant effect instead came from redistribution of a fixed material budget across structural levels. The Physics Experimenter then assembled the methods, tests, results, hypothesis evaluation, and conclusions into a detailed scientific report. The best designs were passed to the 3D Printing Bot. It opened Bambu Studio and brought the Bambu Lab H2D online. Both STLs were placed on one build plate at the same 50x scale and sliced using a 0.20 mm PLA process. The prints completed within less than an hour. The loop images → structural abstraction → executable physics → autonomous experiments → hypothesis testing → design selection → STL → slicing/manufacturing code → physical object That last transition is what I find especially interesting: AI is beginning to operate across the entire scientific and physical workflow - converting observations into models, models into experiments, experimental evidence into revised designs, and those designs into manufactured matter by directly operating machines. This starts to blur the boundary between AI that reasons about the physical world and AI that can actually act on it. Shoutout to the @bot team - you are building something very special here! The way these agents can move naturally from reasoning, to experiments, to operating machines in the physical world feels like an important step.

Can we compile matter - for instance, a pine cone - and derive new active materials, end-to-end from observation to manufacturing? If physical systems can be formalized as composable mathematics, we can point AI that has been shown to resolve long-open mathematical problems at matter itself. Our new work turns bioinspired engineering from analogy into formal compilation: biology and mechanics become explicit, checkable, and executable, so AI reasoning can produce physical designs. This is the first end-to-end demonstration in which a formally compositional multiscale model is carried from a biological hierarchy, through engineered design and fabrication specification, to executable manufacturing code - and then to a physically tested artifact. Background: Humans have long been inspired by biology to advance technology, but this has usually been an ad hoc process rather than a mathematically rigorous one. Natural materials such as pinecones achieve adaptive behavior through mechanisms organized across many scales. Engineering typically translates those mechanisms by analogy: identify a biological principle, build something inspired by it, and validate each new design as a separate case. This can produce remarkable results, but the knowledge does not readily compound. Instead, we represent each scale as a dynamical module with explicit states, stimuli, governing laws, and interfaces. Every scale-to-scale map must preserve the stimulus - response dynamics: evolve the fine-scale system and then map upward, or map upward first and then evolve. The two paths must agree. Because this condition is preserved under composition, locally valid interfaces remain consistent when assembled into the full hierarchy. We then carry that structure into an engineered system, translate the target behavior into a verified fabrication specification, and compile it into G-code: the toolpaths, deposition sequence, temperatures, speeds, and other commands executed by a 3D printer. The intermediate translations are explicit, checkable, and executable rather than completed through an ad hoc handoff. The formal guarantee is that given valid local models and interfaces, their composition remains valid. Whether those models and manufacturing assumptions accurately capture physical reality remains an empirical question. That is why we fabricated and tested the results. We generated four actuator classes by crossing two stimuli - humidity and heat - with two responses: bending and twisting. The fourth, thermal twisting, required no new pipeline and no separate derivation within the framework. It emerged by composing a thermal stimulus module already validated in one case with a twisting module validated in another. The generated G-code produced the intended motion without manual redesign, and all four predictions fell within one experimental standard deviation of the measured response. Why this matters: 1⃣For AI in science, this provides a physics-aware type system against which generative proposals can be checked - and rejected at the interface - before expensive simulation, fabrication, or experiment. It is roughly analogous to proof checking, but for the composition of physical mechanisms. 2⃣For engineering, the accessible design space can scale with a library of validated components rather than with the number of individually derived cases. 3⃣The mathematics, category theory, carries all the way into a physical object on a print bed. This points toward scientific knowledge as executable infrastructure: models that are not only described in papers, but typed, composable, verifiable, and able to compile into experiments. Excellent work led by my student @leemmarom with @SkylarTibbits & @GioeleZardini. Paper published in J. Mech. Phys. Solids along with code, Grasshopper scripts, and manufacturing G-code below.