Diagram: Heliogenesis, 2026

Industry moves one hundred billion tons of material every year, and about ninety percent of that mass is never recycled. We mine a planet to build things that become waste. That is not a failure of the machine. It is the machine.

The usual response is efficiency: mine a little less, recycle a little more, put a price on the carbon. Every one of those moves keeps the same shape, because the input is still something dug out of the ground and the cost is still displaced onto whatever cannot send an invoice. The soil cannot send an invoice. Neither can an aquifer, a pollinator, or a person born in 2070.

A different move is available, and it is no longer speculative. Instead of extracting a material and shaping it, you grow it. Bacteria precipitate calcium carbonate and give you cement without a kiln. Fungus eats agricultural residue and gives you a finished panel in a single week. Bacteria fed on methane give you a plastic that is carbon negative. In every case the input is biology and sunlight, and the factory is a tank rather than a quarry, standing in an ordinary industrial building.

We catalogued the field to find out how far it has actually got. One hundred and seventy-six projects. One hundred and four of them already sell at commercial scale. The result surprised us, and it should change the argument entirely.

One hundred and forty entries fall under material independence. Thirty-two use bacterial growth or fermentation. Twenty-seven work at molecular scale. Twenty-three build materials that stay alive after they ship. Twelve use algae or seaweed, twelve use fast-growing plant biomass, nine work on engineered wood, seven grow mycelium. Four grow cells in a tank, and six draw on indigenous and traditional practice. Only six recover value from a waste stream, which is the clearest gap in the whole set and the one we intend to fund first.

Geography tells its own story. Europe holds seventy-seven of the projects and North America seventy. Asia holds seventeen, Oceania seven, Latin America four, and the entire African continent holds one. A technology that claims to make a region independent of imported material is, at the moment, developed almost entirely in the two regions that already import the least of it. That imbalance is not a detail. It decides who owns the specification when the technology finally arrives at scale.

Maturity is the number that changes the conversation. One hundred and four entries sell commercially, eleven are early commercial, thirty-five run as pilots, and twenty-three sit in a laboratory. Two have paused or restructured, which we record rather than hide. The science question is largely settled. What remains is an industrial and a political question. That distinction matters more than any number.

This is why we publish the catalogue rather than a manifesto. A manifesto asks you to believe something. A catalogue asks you to check it, and it names the firm, the founder, the location, the funding, and the scale, so that checking is cheap. Where a number is undisclosed we write undisclosed. Where a project has stalled we say so, because a list that only contains good news is not a list, it is an advertisement. An advertisement is easy to write, and nobody can check it.

A better material changes little if the same owner captures the gain, so material independence is only one of three vectors. Seventeen entries cover community coordination: Mondragon, the cooperative district of Emilia-Romagna, the Preston procurement model, the Sardex credit circuit, the WIR bank trading since 1934. None of these is a proposal. All of them are audited, and several are older than the problem. The oldest has traded for ninety years already.

The third vector is the one that people find easiest to dismiss and hardest to argue with. A machine pointed at the wrong goal will reach it. Nineteen entries show a state treating inner development as public infrastructure rather than as a private hobby: forest therapy sites in Japan, free nature prescriptions in New Zealand since 2025, nine wellbeing domains screening every policy in Bhutan, outdoor education written into the Finnish curriculum. Fifteen of the nineteen are permanent programmes. We did not expect that, and it changed the way we now write about the third vector entirely.

Our own exit model belongs in the same argument. Heliogenesis runs on Exit-to-Planet, which means every specification, dataset, and piece of intellectual property is released into the commons instead of sold. This is not generosity, and it is not a marketing line. A transition that depends on one owner staying benevolent for forty years is a bet, and the historical record of that particular bet is bad enough that we would rather remove it from the design entirely. We would rather hand a specification to a competitor today than hold it for a valuation that arrives late and helps nobody.

Phase two now turns the catalogue into signed agreements. Ten projects are marked as top candidates and twenty-two more as a strong fit. The test is signatures on paper, not meetings.

Back to all work