Writing is one way to test an idea. Building something people use is another.
Our
work
The essays argue. These projects try. If what we write is right, it should hold up when real people use something built on it, in a conversation, in a community or inside an organisation. If it doesn’t hold up, we would rather find out here than keep writing as if it did.
Earth & Us
Sustainability begins with human understanding. Earth & Us helps organizations examine their understanding, conduct and consequences, improve their practices, and work toward independently verified sustainability.
What it’s testing: whether sustainability can be assessed across the whole organization, understanding included, with improvement support kept apart from the certification decision.
Read about Earth & Us → CompanionDhee
A companion for questions about life.
What it’s testing: whether a conversation can end in understanding rather than in endless engagement.
dhee.co → CommunityCoSeeker
A place to examine our understanding together.
What it’s testing: whether people can say “I don’t know yet” in public without losing standing, and say what a discussion actually resolved.
coseeker.app → OrganisationsK4M2 AI
Agents can analyse. They cannot yet judge. K4M2 AI builds agent systems for organisations around that difference: analysis at machine scale, the organisation’s criteria written down plainly, and commitment kept with its people.
What it’s testing: whether a sense of what matters can be written down clearly enough to guide AI, while people keep the final say.
k4m2.ai → PodcastWhat is K4M2A?
One person, one hour, one question.
What it’s testing: whether talking and listening, again and again, really does lead to better solutions.
k4m2a.org →About money
This website has no ads and sells nothing. K4M2 AI is different: it works with organisations as paid projects. We say so here so you can weigh what we write about technology knowing that. Earth & Us is exploring a voluntary contribution model, so that organizations with more could help smaller ones take part. It is still being designed, and nothing is being collected yet.
How we judge our own work
The same test as everything else on this site: does it help people understand, and does that understanding show in how they live? A project that keeps people busy without helping them understand has failed, however popular it is. Once a year we report what we learned, including what didn’t work.
Where this might be wrong
- We built these projects, so we are not neutral about them.
- Usage numbers can look like success without showing whether anyone understood more. We are still working out how to tell the difference.
- Paid work brings its own pressures. We try to keep them visible rather than pretend they aren’t there.