In Liu Cixin's The Three-Body Problem, the universe is a dark forest.

It is full of intelligent life, but it is dead silent. Every civilization is an armed hunter stalking through the trees. Announcing your existence is a death sentence. The smart ones stay quiet.

For the last decade, the tech company playbook was simple. Build in public. Broadcast your momentum. Turn visibility into customer acquisition. That worked when shipping software was expensive. It worked when the sheer cost of engineering was a barrier to entry.

AI is destroying this barrier. The cost to replicate digital value is approaching zero. What used to take a team of thirty engineers six months now takes a team of three in six weeks. When creation becomes practically free, infinite and quasi-weird (more on this in the future) supply floods the market.

I think we're entering the era of Dark Forest Markets.

A Dark Forest Market is an ecosystem where visibility is a liability. If you stand in a clearing and broadcast high margins in an easily understood software category, you become a target. AI-native startups with zero legacy costs are the hunters. They will see your margins. They will clone your core workflow in weeks. They will gut your pricing power and move on to the next signal.

The Moat of Obscurity

In a world where margins compress by default, the loudest players lose. They compete on dimensions that get cheaper every quarter. The rational response is not to fight the compression. The rational response is to stop broadcasting and start accumulating.

You must find a niche of durable complexity.

You must quietly build a dominant position.

You must resist every temptation to advertise that you are winning.

You are looking for problems where AI and commoditization slow down. The technology might be capable, but the surface area of the problem is too narrow and too contextual to attack at scale. Look for regulatory environments. Look for deep domain workflows. Look for hyper-specific industry integrations. Look for trust relationships built over decades.

A defensible niche has clear properties:
  • The problem is completely invisible to outsiders.
  • The buyer is a specific person at a specific type of company.
  • The solution relies on undocumented institutional knowledge.
  • The market is highly profitable but looks too small to attract venture-backed competitors burning cash for market share.
Boring to an outsider, invaluable to an insider, and small enough to be left alone.

The Signal and the Noise

There is an obvious irony in writing publicly about staying quiet. But the theory is not that silence is universally good. In a dark forest, you must be deliberate about when you make noise. You must control who hears it and what signal you send.

Broadcasting that you found a lucrative niche is noise. Operating in that niche and compounding your position is the actual strategy.

The companies that will look inexplicably valuable five years from now are not making the most noise today. They found a corner, went dark, and got to work.

Niches are riches. The rest is noise.