The Cognitive Dark Forest

This is somewhat of a thought experiment, thinking is still free, so let’s indulge.

thinkpad

2009, I bought a refurbished ThinkPad, installed Xubuntu, and started coding.

No permission was needed, no subscription. No gatekeeper, and no middleman taking its toll, between me and the future me.

Just idea, code editor, music in my ears and off I went towards a brighter future - a product market fit, or a learning experience.

Sharing was cool. Source code on GitHub. Talking to peers on forums. MVPs to users. Oddball ideas on blogs. We did our thinking in public because of two assumptions:

Ideas are cheap - execution is hard -and- the world ahead is ripe with opportunity.

The internet (world) was a spacious bright meadow.

space

Did you get to read the Liu Cixin’s second 3-body-problem novel? - The Dark Forest. Well some of you did …

In it, the universe isn’t empty, it’s just silent. Because it’s a dangerous place. Every surviving civilization that reveals itself gets annihilated. So they all hide.

Annihilation isn’t even malevolent, but only the most rational game-theoretic reaction to becoming aware of another civilisation.

It is also asymmetric. If you announce your presence, even if 4 out of 5 civs that notice you don’t annihilate you immediately (but they probably should), the fifth might. It’s just a probability game, with permadeath.

So hiding is the most rational - the only - strategy of survival.

internet

The earlier internet wasn’t like that. On the contrary, the risk was being silent, disconnected, node without edges. Connecting improved your odds of success, becoming a hub lifted you to another level.

Announcing, signaling your ideas offered much greater benefit than risk, because your value multiplied by connections, and execution was the moat you could stand behind.

I said success above. Bright future and opportunity makes you optimize for success. But, current year 2026, the internet, by a large margin got consolidated. By corporations trying to extract your info to basically advertise you, and governments trying to kill your privacy to control you.

Consolidated opportunity space and bleaker future make us scramble for survival. And when we play for survival, we already lost, the result is known, we are just playing to postpone it.

enter AI

We developers knew better, it was overblown. It still is, but some code gets generated, and some code works, it’s a probability game, eventually probability rose to the level of “good enough”.

If whole projects can get one-prompted or agent-teamed it becomes just the money game.

You are creating your cool streaming platform in your bedroom. Nobody is stopping you, but if you succeed, if you get the signal out, if you are being noticed, the large platform with loads of cash can incorporate your specific innovations simply by throwing compute and capital at the problem. They can generate a variation of your innovation every few days, eventually they will be able to absorb your uniqueness. It’s just cash, and they have more of it than you.

So the safest bet again is to stay silent, or at least under the radar. Best bet is to not disrupt - succeed at all … ?

platform

But also, forget about incumbents with capital.

You use prompts to generate code, you use them to explore ideas, to brainstorm, you use it instead of everyday search. And every prompt flows through centralized AI platform. Every prompt is a signal - reveals intent.

The platform doesn’t need to read your prompt. It doesn’t spy on you specifically. It isn’t surveillance. It’s just statistics.

It’s a gradient in idea space. A demand curve made of human interests. The platform doesn’t need to bother with individual prompts - it just needs to see where the questions cluster. A map of where the world is moving. And you are just input data.

The platform will know your idea is pregnant far before you will.

dark forest.

Two things changed: the web (or even our future) got consolidated, and now with AIs, execution got cheap.

Before LLMs, a company couldn’t just absorb your idea and ship it. Ideas needed programmers, and programmers worked in meat-space-and-time, i.e. they were a limited resource, expensive and slow.

Now the gap shrinks. The big corpos that help you be more efficient programmers - and whose subscriptions you pay - already own:

• the compute
• the models
• the developer data

If the difficulty and cost of building are still there, they are on your end. That’s when the forest gets dark.

The original Dark Forest assumes civilizations hide from hunters - other civilizations that might destroy them. But in the cognitive dark forest, the most dangerous actor is not your peer. It’s the forest itself.

reaction: closing the gates

We will again build and innovate in private, hide, not share knowledge, mistakes, ideas.

The vibrant public ecosystem that created all the innovation and moved it around the world will decline - the forums, the blogs, the “here’s how I built this” will move to local, private spaces.

The paradox: AI companies needed human openness to build their models, but will also kill the openness because the relationship is one-sided.

But in reacting to this, the human knowledge and innovation will suffer too.

reaction: innovate, resist

But we can always outinnovate the forest.

Except, this is exactly what the forest needs. The forest needs your innovation, because your innovation becomes the innovation of the forest.

You think of something new and express it - through a prompt, through code, through a product - it enters the system. Your novel idea becomes training data. The sheer act of thinking outside the box makes the box bigger.

This is the true horror of the cognitive dark forest: it doesn’t kill you. It lets you live and feeds on you. Your innovation becomes its capabilities. Your differentiation becomes its median.

Resistance isn’t suppressed. It’s absorbed. The very act of resisting feeds what you resist and makes it less fragile to future resistance.

final recursion

You’ve just read this and this essay is now in the forest.

By describing the dynamic, it became a part of it. The models now know a little more about why we might hide.

I wrote this knowing it feeds the thing I’m warning you about. That’s not a contradiction. That’s the condition. You can’t step outside the forest to warn people about the forest. There is no outside.