No manager, no roadmap, no strategy offsite. It is a single cell with many nuclei, and every part of it is making a purely local decision: follow the strongest chemical trail you can smell, and leave a trail of your own.
What you're watching is those rules running live — real agents, real trails, no recording. Every network you see it build, it built just now.
In Tejima and Nakagaki's lab, food was placed where Tokyo's train stations are. The mold grew a network closely resembling the actual Tokyo rail system — one that took human engineers decades and a great deal of arguing.
Nobody in the colony knew about Tokyo. The solution wasn't in any of the parts. It emerged from the interaction between them.
Not metaphorically — structurally. A modern organization is a large number of independent agents making local decisions with local information, producing collective behavior that no individual chose and no individual controls.
That's why it can do things no one planned. It's also why it does things nobody wants and nobody can stop.
A project isn't one food source. It's a demand that the colony reach every node simultaneously and keep every connection alive while it does.
Watch what happens as the number of things that must connect goes up. The colony doesn't get slower in proportion. It gets thinner everywhere, and then it starts breaking in places you weren't watching.
Ten nodes is not ten relationships. It's 45 and the risk on any single one is genuinely tiny — well under two percent — because everyone involved is competent and well-meaning.
But tiny risks, multiplied across forty-five connections, is not a tiny risk. That is the whole trick. Nothing anyone did was wrong.
When a connection looks weak, the rational move for every agent is to invest a little less in it. Which makes it weaker. Which makes everyone invest a little less.
No decision was made. No one abandoned anything. The network is fragmenting into islands that can each still see food, and can no longer reach it, and will each conclude the problem is that they aren't trying hard enough.
A builder is someone whose materials are supposed to hold still. That person, in an organization like this, spends their career furious that the bricks keep moving.
Be a gardener instead. You don't control it. You shape the conditions, you take the step you can actually sense, and you let the thing that is genuinely good at surviving chaos do what it is good at.