Most (but not all) of my writings are published here on Less Wrong. You can find links to everything I write by visiting my blog lsusr.com. You can subscribe to my posts via RSS.
lsusr
Karma: 10,194
That’s a good question. I think the answer is “no” because each Western power had lots of rivals.
The Cold War was a different story. In the Cold War, there were (in theory) only two opposing sides. The USA would fund basically anyone who opposed the USSR (and vice versa).
Coordinating the Unequal Treaties
You’re not wrong. Context does indeed matter. Few systems fall perfectly into first-strike vs second-strike.
First Strike and Second Strike
I wanted to give readers the experience of what it was like for me to read the book.
Fixed. Thanks.
[Book Review] “Sorceror’s Apprentice” by Tahir Shah
I agree that GPT-3 sounds like a person on autopilot.
The 1940′s would like to remind you that one does not need nanobots to refine uranium.
I’m confused. Nobody has ever used nanobots to refine uranium.
I’m pretty sure if I had $1 trillion and a functional design for a nuclear ICBM I could work out how to take over the world without any further help from the AI.
Really? How would you do it? The Supreme Leader of North Korea has basically those resources and has utterly failed to conquer South Korea, much less the whole world. Israel and Iran are in similar situations and they’re mere regional powers.
Designing nuclear weapons isn’t any use. The limiting factor in manufacturing nuclear weapons is uranium and industrial capacity, not technical know-how. That (I presume) is why Eliezer cares about nanobots. Self-replicating nanobots can plausibly create a greater power differential at a lower physical capital investment.
Do I think that the simplest AI capable of taking over the world (for practical purposes) can’t be boxed if it doesn’t want to be boxed? I’m not sure. I think that is a slightly different from whether an AI fooms straight from 1 to 2. I think there are many different powerful AI designs. I predict some of them can be boxed. Also, I don’t know how good you are at taking over the world. Some people need to inherit an empire. Around 1200, one guy did it with like a single horse.
Thank you for the quality feedback. As you know, I have a high opinion of your work.
I have replaced “outer alignment” with “bad actor risk”. Thank you for the correction.
The way I look at things, an AGI fooms straight from 1 to 2. At that point it has subdued all competing intelligences and can take it’s time getting to 3. I don’t think 2 can plausibly be boxed.
I played American football for two years. It was a lot of fun.
I had a online friend I made through foreign language learning provide a source of KN95 masks at the height of the COVID-19 shortage. He lives under an authoritarian government. Long-term relationships are one way how you avoid scams over there.
Welcome!
[T]he many-worlds interpretation of quantum mechanics. Such a view would completely destroy the idea that this world is the special creation of an Omni-Max God who has carefully been steering Earth history as part of His Grand Design.
One planet. A hundred billion souls. Four thousand years. Such small ambitions for an ultimate being of infinite power like Vishnu, Shiva or Yahweh. It seems more appropriately scoped for a minor deity.
Definition implies equality. Equality is commutative. If “human values” equals “whatever vague cluster of things human brains are pointing at” then “whatever vague cluster of things human brains are pointing at” equals “human values”.
I hope the $10 in cryptocurrency I get for saving energy is proof of work. I am ideologically opposed to proof of stake.
I appreciate the charity for machine rights. Machines are people too.
I want to hack someone else’s neuro-pellets and Rickroll them.
Human brains are a priori aligned with human values. Human brains are proof positive that a general intelligence can be aligned with human values. Wetware is an awful computational substrate. Silicon ought to work better.
Re: Attempted Gears Analysis of AGI Intervention Discussion With Eliezer
I read your Hacker News post. What don’t you like about the curriculum? If the answer is “it’s too easy” or “I hate Java” then you should take seriously the idea of dropping out (or if you’re a freshman then consider changing your major to something harder like math or physics). If the classes aren’t hard enough then the biggest thing you (personally) will lose if you drop out of college is an easy entry ticket into the big tech firms like Amazon, Facebook, etcetera. Try to arrange for a company to hire you early, before you graduate. If you succeed then you can drop out with basically no risk. If you fail to secure an early job then you will have spent a bit of time learning useful things about how the world works.
What do you want to be doing in 1 year after graduation? 2 years? 5 years? 10?
Does your offer include annotating your thoughts too or does it only include writing the prompts?