A game theory behind the dark forest strategy? | Hacker News Reader

A game theory behind the dark forest strategy?

en.wikipedia.org

9 pointsdgregd13 comments

What do you think about the dark forest hypothesis from The Three-Body Problem book? It seems reasonable to me. Are there any sociologist or math papers that describe the game theory behind the dark forest strategy and prove it?

I'm not very impressed. In general, I think that the people who have hypothesized about the fermi paradox and the natural paths of civilizations have tended to dramatically overestimated the actual detectability of other civilizations, and the feasible growth ceiling of other civilizations by several orders of magnitude.

We could not detect ourselves even in a nearby solar system. And as we've gotten more developed, we tend to be MORE frugal with our energy usage.

One major weakness of the Dark Forest hypothesis is that it presumes the bad guys have had interstellar travel for a long long time ... and yet in all that time they have never come to our system and left any traces behind. They didn't do anything bad to us for 4 billion years, but now they're going to do bad things because we have radio?

space is big. there are probably rocks within sight of I-95 that have never been touched by a human, yet tens of thousands of humans flow by them every day. We could be in the middle of a galactic empire and not know it because our system is a little off the trade lanes and not worth visiting on its own.

For me, the Dark Forest hypothesis is very probable. However, the whole book series assumes that interstellar travel is possible for advanced civilizations. What if no new physics is discovered, and the only way to accelerate/decelerate is to throw a mass in the opposite direction? Maybe all advanced civilizations in the Milky Way are limited to their planetary systems.

There are two ways to go to the stars: go faster or live longer (or hibernate).

If you could cryosleep, lived for millions of years, or were an AI that could just turn itself off for a while, you could go to the stars using simple chemical propulsion. No physicists’ nightmare torch rockets or warp drives required.

The crazy physics requirement is to enable interstellar flight within natural human life spans. It’s probably easier to extend that or figure out how to hibernate or become AI than it is to build an antimatter rocket or bend space time (if that is even possible).

I doubt the dark forest because Earth has been broadcasting that it’s a likely biosphere via its albedo absorption spectrum for over a billion years. If you are a hyper paranoid alien you should just blast every biosphere to kill any possible rival before it even evolves.

Side issue, but if you don't care about travel time, you also have another big problem. The frequency with which stuff happens that matters to civilizations has increased significantly over time. Meaning that the time intervals that matter have gone down a LOT as we've evolved and that seems like a natural tendency on the basis of natural selection: higher frequency strategies will tend to dominate. So then you have this massive juxtaposition between probe travel times and things civilizations care about. It's a total mismatch. By the time the probe reaches its destination and can come back, the civilization will likely have completely changed and/or forgotten about it.

This is true for us, but is also anthropocentric. We can't assume that everyone would have this problem.

Technological change is probably a sigmoid like most things. There could be people out there who have already won physics, figured out the entire unified theory of how everything works, and have developed most of the core technological stuff they ever will. At that point they might calm back down and become more stable over long periods of time. These are the kinds of civilizations that would send slow probes or one-way colony ships out and would be willing and able to wait.

It's very hard to think about potential aliens without dragging in anthropocentric or even recent-history-centric assumptions.

But we're also not a threat - even if we did detect another civilization, we'd have no way to attack them.

I don't buy the Dark Forest hypothesis at all, but it's also just not particularly a factor for us at our current stage of technological development, any more than a stone-age tribe has to worry about Mutually Assured Destruction via nuclear ICMBs.

On the other hand, better to do the weeding early than wait for it to get out of hand...

With that sort of logic, you might as well exterminate every solar system around, whether or not it shows signs of civilization or even eventual habitability.

The last point is critical. Maybe super advanced civilizations actually become so efficient that they use less energy than we do now.

It seems to me that any technologically advanced civilization that is capable of completely annihilating another at such distances would likely be equally capable of avoiding detection, thereby reducing the potential threat level posed by the less advanced civilization. Avoidance would seem to be the best strategy unless the civilization posed a direct and imminent threat. Otherwise, the more advanced beings risk exposing themselves (by deploying a weapon) and declaring to the universe that they are both hostile and aggressive to others.

Reply on news.ycombinator.com