carado.moe

tammy's blog about
AI alignment, utopia, anthropics, and more;

my best alignment plan: QACI!

also on twitter, lesswrong, rss

hope for infinite compute

here are some reasons we may have infinite universe to inhabit in the future.

• encoding ourselves in heat death noise. this could at least buy us exponentially much time to exist; it becomes infinite if the amount of possible states also increases.
• where are we in the universal distribution ? most models that produce rich computation, such as rule 30, seem to grow forever in amount of stuff. this includes wolfram's hypergraph rewriting system.
• space is expanding. the planck length is a constant. unless there's something i'm mistaken about, this sure seems like more positions in space we could inhabit, and thus an increasing total amount of states the universe can be in. it is not obvious how to utilize this, but it may be evidence for other ways in which the amount of stuff in the universe grows. in wolfram's perspective, it is simply the hypergraph creating new positions in space, as it has been doing forever.
• why are there 10⁸⁰ particles in the observable universe? if the total number is larger, why is it that larger number? wouldn't it be occam-simpler that there be 1 or few particles (or qubits or whatever) in the start, and have that amount grow over time? in fact, with expanding space, won't there statistically tend to be more particles overall, if only because there's more space for quantum fluctuations to randomly spawn particles? surely a superintelligence can harness this.
• even if we're in a physically finite universe, we may be able to acausally trade/hack/blackmail aliens living in infinite worlds such as rule 30. maybe. acausal trading is weird.