Consider that “if AGI is very near” probably means that it’s already happened (or, equivalently, that we are past the point of no return) on Copernican grounds, since the odds of living in a very special moment where the timelines are short but it’s not too late yet are very low. Not seeing an obvious AGI around likely means that either it’s not very near, or that the take-off is slow, not fast.
Ironically, it’s not Roko’s basilisk that is an infohazard, it’s the “AGI go foom!” idea that is.
It’s tempting to try to reinvent the wheel, but this dynamic is by no means new. There have been viable political alternatives popping in the middle in various places around the world. Not as many as those emerging from the right or from the left, One can argue that the US is unique in many ways, and it sure is, but the degree of uniqueness would only become clear once you identify the common trends.
From what I understand, the process of emergence of a centrist party is usually by one of the mainstream parties not being radical enough for a large chunk of its base, splitting the party in two, one more extreme and one more centrist. It happened in Canada, Germany, Israel, Italy and many other places. The odds of creating a centrist political force from scratch are not good, and require much shallow equilibria than those in most de facto two-party systems. For example, the Israel Resilience Party was created in 2018 on the multi-party background and many years of political gridlock.
Despite this, superrational reasoning gives us different results.
what is that “superrational” reasoning that gives different results?
One reference point: in Canada the guidance is that the first dose provides enough protection to delay the second dose up to 4 month.
If you appear to be an outlier, it’s worth investigating why precisely, instead of stopping at one observation and trying to make sense of it using essentially an outside view. There are generally higher-probability models in the inside view, such as “I have hallucinated other people dying/playing” or “I always end up with an empty barrel”
Hmm, it sort of makes sense, but possible_world_augmented() returns not just a set of worlds, but a set of pairs, (world, probability). For example for the transparent Newcomb’s you get possible_world_augmented() returns {(<1-box, million>, 1), (<2-box, thousand>, 0)}. And that’s enough to calculate EV, and conclude which “decision” (i.e. possible_world_augmented() given decision X) results in maxEV. Come to think of it, if you tabulate this, you end up with what I talked about in that post.
I’m confused… What you call the “Pure Reality” view seems to work just fine, no? (I think you had a different name for it, pure counterfactuals or something.) What do you need counterfactuals/Augmented Reality for? Presumably making decisions thanks to “having a choice” in this framework, right? In the pure reality framework the “student and the test” example one would dispassionately calculate what kind of a student algorithm passes the test, without talking about making a decision to study or not to study. Same with the Newcomb’s, of course, one just looks at what kind of agents end up with a given payoff. So… why pick an AR view over the PR view, what’s the benefit?
“look through this collection of psychology research and take it as roughly true”
Well, you are an intelligence that “is well-grounded and understands what human concepts mean”, do you think that the above approach would lead you to distill the right assumptions?
Total of 39 cases of SA variant seems too underpowered to make any conclusions.
Note that everything on your list has zero or near zero replication cost. A lot of essentials, tangible and intangible, are not like that. Food, companionship, living accommodations, etc. I don’t know how far one can get on easy-clone stuff.
The pharma research company appears to be very real https://sanotize.com/covid-19/
An important consideration is what you are dealing with. Severe ACE? Something else?
Well, we may have had this argument before, likely more than once, so probably no point rehashing it. I appreciate you expressing your views succinctly though.
Well, it looks like you declare “outperforming” by your own metric, not by anything generally accepted.
(Also, I take issue with the last two. The philosophical ideas about time are generally not about time, but about “time”, i.e. about how humans perceive and understand passage of time. So distinguishing between A and B is about humans, not about time, unlike, say, Special and General Relativity, which provide a useful model of time and spacetime.
A non-epistemic theory of truth (e.g. there is an objective truth we try to learn) is detrimental in general, because it inevitably deteriorates into debates about untestables, like other branches of a hypothetical multiverse and how to behave morally in an infinite universe.)
Also, most people here, while giving lip service to non-libertarian views of free will, sneak it in anyway, as evidenced by relying on “free choice” in nearly all decision theory discussions.
This sounds like a very Eliezer-like approach: “I don’t have to convince you, a professional who spent decades learning and researching the subject matter, here is the truth, throw away your old culture and learn from me, even though I never bothered to learn what you learned!” While there are certainly plenty of cases where this is valid, in any kind of evidence-based sciences the odds of it being successful are slim to none (the infamous QM sequence is one example of a failed foray like that. Well, maybe not failed, just uninteresting). I want to agree with you on the philosophy of religion, of course, because, well, if you start with a failed premise, you can spend all your life analyzing noise, like the writers of Talmud did. But an outside view says that the Chesterton fence of an existing academic culture is there for a reason, including the philosophical traditions dating back millennia.
An SSC-like approach seems much more reliable in terms of advancing a particular field. Scott spends inordinate amount of time understanding the existing fences, how they came to be and why they are there still, before advancing an argument why it might be a good idea to move them, and how to test if the move is good. I think that leads to him being taken much more seriously by the professionals in the area he writes about.
I gather that both approaches have merit, as there is generally no arguing with someone who is in a “diseased discipline”, but one has to be very careful affixing that label on the whole field of research, even if it seems obvious to an outsider. Or to an insider, if you follow the debates about whether the String Theory is a diseased field in physics.
Still, except for the super-geniuses among us, it is much safer to understand the ins and outs before declaring that the giga-IQ-hours spent by humanity on a given topic are a waste or a dead end. The jury is still out on whether Eliezer and MIRI in general qualify.
These are some extraordinary claims. I wonder if there is a metric that mainstream analytical philosophers would agree to use to evaluate statements like
LW outperform analytic philosophy
and
LW is academic philosophy, rebooted with better people than Plato as its Pater Patriae.
Without an agreed upon evaluation criteria, this is just tooting one’s own horn, wouldn’t you agree?
You and I rarely agree on much, but this looks like a great post! It highlights what an outsider like me would find befuddling about philosophical discourse, and your prescriptions, usually the weakest part of any argument, actually make sense. Huh.
The link in the reply requires an iPhone app, no Android or desktop support. That seems a bit limiting.
Huh, what an interesting observation! I wonder if there are more examples of something like that.