Neri Oxman on Material Ecology, Designing with Nature, and Engineering with Biology

Neri Oxman is a designer, architect, and scientist who founded the Mediated Matter group at the MIT Media Lab and now runs Oxman, a company built on a single ambition: to grow products the way nature grows organisms, rather than assemble them from inert, toxic parts. Across nearly three hours she lays out the field she named in 2005 — Material Ecology — and walks Lex through the pavilions, death masks, and space-flown bees that test it, before turning to beauty, faith, love, extinction, and what nature might want if it could be asked.

Key ideas

1. Human-made stuff now outweighs all living things. Drawing on the Weizmann Institute’s Ron Milo, Oxman fixes 2020 as the crossover year when anthropomass — everything humankind has built, from iPhones and concrete to asphalt and plastic — first exceeded the planet’s total biomass, every living organism combined. That imbalance is the problem her company exists to address: not less building, but building that gives back.

2. Material Ecology: design everything as if it were grown. The field she coined treats every material object as part of the ecology, written and read as nature is. Its informal test: could you grow an iPhone, a car, a building, so that owning one heals nature rather than depletes it? The slogan is ‘grow everything’, and the goal is a world where you cannot tell the grown from the made — and it no longer matters.

3. Co-fabrication with hero organisms, guided by templates. Rather than betting on a single ‘hero organism’ to solve everything, her group enlists many — silkworms, bees, bacteria, shrimp shells — and guides their building through computational templating: physical, environmental, chemical, and geometric scaffolds that let nature and machine ‘duet’. 17,532 silkworms spun the Silk Pavilion; E. coli pigmented the Vespers death masks at twenty microns per voxel.

4. From templating to emergence — and ethical limits. The aim is to template until templating is no longer needed, the moment a biological system ‘assumes agency’ and the swarm-like result is genuine, not authored. (‘If you can predict, it doesn’t count as emergence.’) The work is win-win by design: she refused transgenic silkworms but works freely with bacteria, drawing a deliberate ethical range across the kingdoms of life.

5. Giving nature the bandwidth humans already have. Oxman wants ‘large molecule models’ to decode the chemical language plants and microbes use — green leaf volatiles in cut grass, a rose’s different molecular footprint by night and day — so that nature could access something like the cloud. The first product is a ‘functionalised fragrance’, a message in a bottle that empowers a rose garden rather than merely smelling nice.

Content

Anthropomass overtakes biomass

The episode’s spine is a single statistic. In 2020, Oxman recounts, the Weizmann Institute’s Ron Milo identified the crossover year in which anthropomass — the combined mass of everything humankind has designed and produced — first exceeded the planet’s biomass, the combined mass of all living organisms. Plastics, wearables, buildings, cities, asphalt, concrete: the made now outweighs the living. ‘More phones than bones’, as the phrase goes; the technosphere outweighing the biosphere. A Zoom call with Milo became one of those handful of life-moments that ‘get you to course correct’. Her response is not retreat but redirection: a world in which driving a car could be better for nature than a world with no cars, in which buildings augment and heal rather than merely occupy.

Material Ecology

In 2005 Oxman named the field that organises all of this: Material Ecology — the idea that every material object can be considered part of the ecology, designed to have a positive impact on it, where ‘all things nature, all things human’ work together. The informal benchmark she calls a Turing test for the field: a material object you could read and write to as nature-grown, so that the grown and the made become indistinguishable and the distinction stops mattering. Nature, in her framing, is ‘everything that isn’t anthropomass’ — and it holds a wisdom that she takes to sit above mere intelligence, a wisdom technology might tap rather than overwrite.

Growing products, not assembling them

Her company, Oxman, works at three scales. At the molecular scale it seeks a universal language of nature and a tool she calls ‘the iPhone for nature’. At the product scale it is building a single object that begins as CO₂ — captured via carbon-recycling technology starting from methane or wastewater — becomes a product that is worn and used, then returns to the soil and grows an edible fruit plant. The world’s first entirely biodegradable, biocompatible, bio-renewable product. Her quarrel with existing ‘bio-based’ efforts is that they solve one link in the chain — mycelium leather, cellulose textile — then fall back on glue and inert parts to finish the shoe. Oxman wants the whole chain, end to end, made by robots and microbes alone.

Synthetic biology — engineering living cells to manufacture useful molecules, the way a brewer’s yeast is redirected to make a drug instead of beer — is the enabling craft throughout. Chitosan, one of her recurring materials, is a biodegradable substance derived from the chitin in shrimp and crab shells. Melanin is the same pigment that colours human skin, here grown and templated by bacteria. None of these is assembled; all are grown.

Computational templating and the hero organisms

Over a decade at MIT, the Mediated Matter group produced what Oxman frankly calls ‘N of ones’ — singular embodiments bound for museums and Nature covers alike (‘it should be at MoMA and the cover of Nature at the same time’). The unifying method was computational templating: building scaffolds — physical (water-soluble meshes), environmental (a robot varying heat and light), chemical (pheromones), geometric — that guide a living ‘hero organism’ as it builds.

The Silk Pavilion is the worked example. Six metres tall, spun by 17,532 silkworms, each thread about a mile long, it now sits at the Museum of Modern Art. The breakthrough was behavioural: a silkworm placed on a flat surface under 21 millimetres spins a flat patch rather than a cocoon, for want of a vertical post to anchor to. The team exploited that to build a dome from flat patches. Silkworms have ‘zero communication’ across individuals — a human-authored species, bred 4,000 years ago to serve textiles, unable even to fly. Yet the robotic templates inadvertently let them cooperate, rubbing against hotter and colder regions, producing a pavilion that ‘seems like it was produced by a swarm’ though no swarm exists. The deeper aim is to move past templating altogether toward emergence — the point at which the organism ‘owns the code’ — captured in her line that a result you can predict ‘doesn’t count as emergence’.

Engineering with bees and bacteria

The Synthetic Apiary began as an architectural project: a perpetual-spring environment for Massachusetts bees, who otherwise lose eighty per cent of their number to winter, proving they could stay reproductive and alive year-round. Its principle — ‘before you can design with, you have to design for’ — became a product when the team sent a queen and her retinue to space aboard a Blue Origin New Shepard flight, outdoing a 1982 NASA experiment in which the bees returned non-reproductive. Oxman’s bees came back alive and kicking, and a Nature paper followed: by feeding them gold and silver nanoparticles, the team proved for the first time that bees recycle wax.

The Vespers death masks pushed bacteria to finer resolution. Fifteen masks — past, present, future — used E. coli to grow pigment patterns; chemical signals embedded in the printer’s resin interacted with pigment-producing bacteria at the surface, computing exact patterns at twenty microns per voxel. The project birthed a material class her student Rachael named Hybrid Living Materials (HLMs), which led on to the melanin-templated Mandela Pavilion. Here the ethics sharpen: invited to make glow-in-the-dark silk by splicing spider genes into silkworms — a transgenic operation — she said no. Silkworms she will not modify genetically; bacteria she will. ‘Not that I’m organism agnostic’, but the range is deliberate, and the rule is win-win: work with an organism only when it is better for the organism too.

Talking to plants

Plants already communicate, Oxman insists — the smell of freshly cut grass is a distress signal, green leaf volatiles warning neighbouring blades. ‘We use language models, they use molecular models.’ The frontier is decoding those molecules: plant neurobiology is real, plants have something akin to a nervous system operating on a far slower timescale, and a rose’s molecular footprint differs by night and day along its circadian rhythm. Read that language and precision agriculture transforms — today’s drone photographs of failing crops arrive ‘already too late’, whereas a molecular footprint predicts distress before it shows. Her first molecular product is a functionalised fragrance: not merely a scent for human indulgence but a ‘message in a bottle’ given agency, able to inspire — never ‘trick’ — a rose garden, while sparing the 10,000 rose bushes a conventional 5 ml of fragrance consumes. System-level thinking, she notes, ‘is almost always going to take you to the entire earth’.

The grow-rooms and the long timescale

The new lab, under construction since 2019, is built around capsules that are simultaneously wet lab and grow-room: top-down environmental templating (light, humidity, temperature) over bottom-up genetic regulation. ‘Ecology in a box’, stressable to produce the food, products, or construction materials of the future — one capsule growing transparent wood, another a shoe, another a biodiversity chamber that recreates ‘Ohio, 1981’ to ask what a tomato tasted like. Designing in the scale of thousands of years — the timescale of redwoods — does not frighten her. The hard part is reconciling that patience with a Western economy ‘based on capitalism’ that needs products now. What we fear from AGI, she wants for biology: ‘What we don’t want to happen with AGI, we want to happen with synthetic biology.‘

Beauty, faith, and what nature wants

The conversation’s second half is philosophical. Beauty, for Oxman, is agency — she invokes Buckminster Fuller: when a finished solution is not beautiful, you know it was wrong. Empowerment she defines, via a colleague’s equation, as a force with direction; emergence is multi-directional, and empowerment can inspire it but not the reverse. Imperfection is necessary — ‘change is the only permanence’ — and a flaw, she suggests, is ‘an increased surface area’, the very thing that lets mortar bind bricks into a home and vulnerability bind humans into community. Threaded through are the reading of Einstein’s letter to his daughter naming love as the universal force, Viktor Frankl on meaning, and her own credo of suspending disbelief and ‘rebooting’ to stay creative. Asked about AGI she is ‘hopeful and terrified’ both: the end of human agency, she ventures, might be the beginning of agency for nature — language models uploaded to a ‘long-lived, humanlike, sentient weeping willow’.

Related

• Michael Levin on Bioelectricity, Basal Cognition, and Diverse Intelligence — Oxman cites Levin directly on ‘gentle material’ that has agency; shared theme of intelligence distributed across living systems
• Sara Walker on the Physics of Life, Time, Complexity and Aliens — complementary inquiry into what life is and how it is defined across timescales
• Lee Cronin on Assembly Theory, the Origin of Life, and Alien Detection — overlapping ground on the origin of life and growing complexity from chemistry
• Joscha Bach on Life, Intelligence, Consciousness, AI and the Future of Humans — Oxman references this conversation on stochastic versus deterministic processes
• Bob Baxley on Design as a Moral Obligation, Design Tenets, and the Primal Mark — a contrasting account of design’s purpose and ethics
• Michael Levin — biologist whose work on agency in living matter Oxman draws on
• Lex Fridman — host