Reddit reviews Anthropic Bias (Studies in Philosophy)

Well we can't use our existence as evidence regarding the prior probability that humans would exist up until now because of anthropic bias. If humans had gone extinct, we wouldn't observe it. So it is plausible that the chance of human extinction (or failure to evolve) was very high. You may want to read https://www.amazon.com/Anthropic-Bias-Observation-Selection-Philosophy/dp/0415883946.

>Given the natural inferiority of our bodies in comparison with stronger potential predators,

That's not really how it works, evolutionary competition is not a one-dimensional scale of strength... chimpanzees, bonobos, and orangutans are weaker than other 'potential predators' but they survive just fine, because their survival does not depend on being stronger. You seem to think that without the modern mind, we wouldn't be able to survive at all, but I can't think of any reason to believe that. Just because we need the modern mind in order to dominate the planet doesn't mean that we need it to survive.

Were early hominids unfit for survival? Well the evidence says no, because they did survive. This wasn't a single probabilistic event, it is clear that they lived in large numbers over a very long period of time and that necessarily implies that they were fit for the environmental conditions that they faced.

> one is in THIS one

But "this" one only means the one you're in. No matter which one you were in, that would be "this" one to you.

If you draw from an urn of poker chips while blindfolded, and there are many more white chips than red, if you have to make a guess whether you're holding a white chip or a red one, white is the best guess. It doesn't mean you know for sure or you've proven it, just that it's the best assumption to make.

To extend the analogy back to Bostrom's argument, he's saying that white chips, signifying simulated universes, assuming for the sake of argument they are possible, would be great in number. He is basing that assessment on our own experience with the utility and popularity of simulation and simulated toy worlds.

>And we don't know how many others simulated or not there are

We do not. Bostrom's reasoning is that, if a civilization develops the technology to run simulations, they would not run it merely in one instance, no more than we developed Virtual Machine frameworks only to run one VM instance. They didn't make Sim City and run the program once. So, his arguments hold, either simulation is impossible, or simulations will be great in number. He doesn't purport to prove that simulating a universe is possible, but he's saying, if you accept these premises, then...


>if we fail to make any simulations this will not mean we know this is real and not simulated.

We already have simulations, just simplified ones. The point is not that we will "know" that our universe is simulated vs real, either way, rather what it is reasonable to assume given the premises.

>that makes no difference to our knowledge of the number of other universes which may or may not exist.

If we manage to simulate universes, that will be proof that doing so is not impossible. We will have established that it can be done. We won't have proven that other technological civilizations exist, or that they engage in universe-simulation, but proving it isn't the point. This is not a scientific hypothesis. Bostrom is a philosopher, and he's exploring reasoning first explored by Brandon Carter in his doomsday argument. It's also an extension of ideas Bostrom explored in his book Anthropic Bias.

Your specific question isn't related because it seems to be due to a confusion, but more generally yes it is related in that there is a lot of work being done in trying to understand how to connect the Born rule to how you "count" the number of worlds. If you really want to dive into this stuff, I can recommend two resources:

Anthropic Bias, by Bostrom (or download legally here)

Everett's thesis (you can also find the thesis online, although this book includes some good commentary)