This year, many of us watched something new happen in software. AI systems can now write code, test it, see what breaks and try again, over and over, faster than any person can follow. What makes that possible is not intelligence alone. It is a feedback loop: a quick and honest answer to the question "did that work?". A programmer has the compiler. A bookkeeper has the ledger. Give a system a good answer to that question, and it improves on its own.
That is exactly what worries people, and rightly so. Feedback loops are powerful, and they optimise hard against whatever they can measure. Today, what they can measure about living nature is close to nothing. So money, policy and increasingly software chase proxies (tonnes of carbon, trees planted, hectares restored), and the proxy gets delivered without the thing itself: a eucalyptus plantation for the carbon, a dead watershed for free.
Feedback loops that work for life
We want those same feedback loops working for life. We want systems that learn and grow along with living ecosystems, where every attempt to restore a place gets an honest, affordable answer: is life here actually coming back? Once that answer is cheap, the twelve-hectare farmer becomes fundable, checking becomes continuous, and the pressure that now pulls the economy away from life can start pulling it towards regeneration.
For that, a system has to be able to sense. That is where the Life Box comes in.
A sensor in the living system
The Life Box is one small, solar-powered box on one post, at about knee height, in a field, a food forest or an agroforestry alley. It listens: a small computer inside recognises bird song on the spot, so only the detections leave the field, never the recordings. Two probes in the ground read soil moisture and temperature. A 4G modem sends it all out on a ten-year IoT SIM, so the landowner's own network is never involved. A solar panel above the box keeps it running.
What the box hears and feels is combined with a third witness that costs nothing: free satellite images from the European Sentinel-2 mission, available for any place on Earth. Three streams that cannot compare notes. Together they give seven readings of what life is doing in that place: productivity, diversity, structure, renewal, cycling, resilience and autonomy. The verdict is set by the weakest reading, so a field that grows lots of green biomass but has gone silent does not count as thriving. When the streams disagree, that is the moment to send a person to go and look.
The skylark in the picture is not decoration. It is exactly what the box is there to notice.
Open, and cheap on purpose
The parts come to roughly €800, with no subscription. The whole thing is a commons: open standard, open code, open hardware and open data, published at life.oncra.org, so anyone can rebuild a box, rerun a verdict or improve the method. The evidence belongs to the place.
This connects directly to our work with Oncra. First, life readings can act as a check on carbon credits: a project should not only store carbon, it should make the place more alive. Later, the same readings could underpin payments for restored nature and the monitoring methods governments are now writing for nature restoration and soil health.
Where we are
The picture above is an impression of the design, not a photograph. The first node has been designed and its parts ordered; the build plan, including every number we still need to measure, is public. Next comes a box on a bench with a meter on it, then a box on a post, then two weeks in the field without anyone touching it.
Do you manage land and want to host one of the first Life Boxes, or do you work on soil, sound or satellite data and want to help? Reply to our newsletter or write to act@climatecleanup.org. We read every email.