So from today's perspective, Friendship is Optimal is a story of someone starting GPT-9 with a silly prompt?

He envisaged masses of people committing suicide because they could not bear the thought of their unhappy lives repeating eternally.

Wait, does this even makes sense? If life repeats eternally, then suicide is a restart of the cycle.

Killing yourself (if you believe in the eternal return, but are otherwise rational) is not about how happy or unhappy you are, but whether you expect the following years to be better or worse than your life until now. So the people killing themselves should be exactly the ones whom Nietzsche admires, when they feel they have already passed their peak creative genius.

Nietzsche certainly respected individual greatness, but I believe his highest respect was reserved for the creators of new value systems, and the leaders and conquerors who made those value systems into the basis of a successful religion or civilization. (...) He saw himself as a philosopher who had imbibed the moral emptiness of a universe of warring wills to power, and summoned up from his own will to power, a new tablet of values.

Guy designed a new moral system putting himself on the top; news at 11.

Nietzsche is popularly associated with Nazism (...) he was also popular among the left-anarchists and the Left generally (...) “if you meet an intellectual non-Leftist, increasingly they are Nietzschean” (whatever that means). Common sense demands that some of these people are misreading him.

Alternative explanation: all these people understand Nietzsche more or less correctly as saying that geniuses are so awesome that nothing else matters, and all of them think of themselves as geniuses.

Seems to me that it shows everything, just like without the filter. It shows my drafts, shortform...

EDIT: Oh, I was wrong. Yes, it shows the drafts and shortform... but in the "Posts" section it only shows the frontpaged posts.

Yes, if there is no deep underlying reason why the two minimal descriptions should be same, and it "just happened", I would assume that with slightly different description language it would not happen.

Even the "3A2B" vs "3ABB" example would stop working if encoding a number used a different number of bits than encoding a character.

Done.

(_) I've made my own post, but never gotten a front page tag

(_) At least one post I made got a front page tag

Is there a quick way to figure this out? When I click on my name, it shows a list of the posts I made, but I need to click on them individually to figure out whether they were at the front page (there is a tag "Frontpage" below the title) or not. Could be difficult for people who made many non-frontpage posts, and don't remember if there was any that made it to the front page.

Yeah, this article is more rational than 95% of politics-related content on LW.

That explains why the NDAs are costly. But if you don't sign one, you can't e.g. get the OpenPhil grant. So the examples don't explain how "it's almost never worth signing one".

When autism was low-status, all you could read was how autism is having a "male brain" and how most autists were males. The dominant paradigm was how autists lack the theory of mind... which nicely matched the stereotype of insensitive and inattentive men.

Now that Twitter culture made autism cool, suddenly there are lots of articles and videos about "overlooked autistic traits in women" (which to me often seem quite the same as the usual autistic traits in men). And the dominant paradigm is how autistic people are actually too sensitive and easily overwhelmed... which nicely matches the stereotype of sensitive women.

For example: difficulty in romantic relationships, difficulty understanding things because you interpret other people's speech literally, anxiety from pretending to be something you are not, suppressing your feelings to make other people comfortable, changing your language and body language to mirror others, being labeled "sensitive" or "gifted", feeling depleted after social events, stimming, being more comfortable in writing than in person, sometimes taking a leadership role because it is easier than being a member of the herd, good at gaslighting yourself, rich inner speech you have trouble articulating, hanging out with people of the opposite sex because you don't do things stereotypical for your gender, excelling at school, awkward at flirting -- haha, nope, definitely couldn't happen to someone like me. /s

(The only point in that video that did not apply symmetrically was: female special interests are usually more socially acceptable than male special interests. It sounds even more convincing when the author puts computer programming in the list of female special interests, so the male special interests are reduced to... trains.)

I suppose the lesson is that if you want to get some empathy for a group of people, you first need to convince the audience that the group consists of women, or at least that there are many women in that group who deserve special attention. Until that happens, anyone can "explain" the group by saying basically: "they are stupid, duh".

You need to make sure that when something important shows up, (1) the system will clearly recognize that this happened, and (2) the producers will actually be able to abandon the lower-priority task quickly.

I have seen companies trying to implement this, but what often actually happens is that the manager responsible for the lower-priority task just keeps assigning work to the employees anyway. The underlying cause is that the manager's incentives are misaligned with the company goals -- his bonus depends on getting the lower-priority task done. (How would you set up his incentives?)

It is also completely unique, being the shortest possible string. It is so lean that not a single redundant detail remains - otherwise, it would not be the shortest string.

I don't think this is necessarily true. (Though I am not sure about it.) I imagine there could be two different compression strategies that both happen to produce a result of the same length, but cannot be merged.

To use your example, "AAABB" can be compressed to either "3A2B" or "3ABB", both containing 4 characters. Knowing that "2B" and "BB" represent the same thing doesn't allow you to exploit this "redundancy" to further reduce it to one character.

Also, some parts of my body and mind are more important than others -- the exact shape of all my hairs at this moment is a lot of data (not easy to compress, because there is a lot of randomness involved), and in a second the shape will be different anyway, and even if you cut my hair short it would still be "me" (at least I do not experience existential horror whenever I get a haircut). Also not sure if gut flora should be included.

I guess my point is that even relatively useless things can require many bits of information and you actually don't need them, some lossy compression would suffice, but if you overdo it, you get The Fly.

Kinda like mathematicians hated it when the four color theorem was solved by a computer brute-forcing thousands of options. Only imagine that the same thing happens to hundreds of important mathematical problems -- the proper way to solve them becomes to reduced them to a huge by finite number of cases, then throw lots of money at a computer who will handle these cases one by one, producing a "proof" that no human will ever be able to verify directly.

Much better.

So, this could be an abstract at the beginning of the sequence, and the individual articles could approximately provide evidence for sentences in this abstract.

Or you could do it the Eliezer's way, and start with posting the articles that provide evidence for the individual sentences (each article containing its own summary), and only afterwards post an article that ties it all together. This way would allow readers to evaluate each article on its own merits, without being distracted by whether they agree or disagree with the conclusion.

It is possible that you have actually tried to do exactly this, but speaking for myself, I never would have guessed so from reading the original articles.

(Also, if your first article gets downvoted, please pause and reflect on that fact. Either your idea is wrong and readers express disagreement, or it is just really badly written and readers express confusion. In either case, pushing forward is not helpful.)

What you see as a broken system, I see as a system working exactly as intended.

Should we keep any nonsense on LW front page just because the author asked us nicely?

Well, yes. I guess it's more of an... expression of frustration. Like telling the space-lizard-Jesus guy: "Dude, have you ever read the Bible?" You don't expect he did, and yes that is the reason why he says what he says... but you also do not really expect him to read it now.

(Then he asks you for help at publishing his own space Bible.)

Shortly:

1. Post the actual text. Not a link to a link to a download site.
2. Make your idea clear. No joking, no... whatever. The inferential distance is too large, the chance to be understood is already too low. Don't make it worse.
3. Post one idea at a time. If your article contains dozen ideas and I disagree with all of them, I am just going to click the downvote button without an explanation. If your article contains one idea and I disagree with it, I may post a reason why I disagree.
4. Don't post many articles at the same time. Among other things, it sends a clear signal that you are not listening to feedback.
5. This actually should be point zero: Consider the possibility that you might actually be wrong. (From my perspective, this possibility is very high.)

I think you are completely missing the entire point of the AI alignment problem.

The problem is how to make the AI recognize good from evil. Not whether upon recognizing good, the AI should print "good" to output, or smile, or clap its hands. Either reaction is equally okay, and can be improved later. The important part is that AI does not print "good" / smile / clap its hands when it figures out a course of action which would, as a side effect, destroy humankind, or do something otherwise horrible (the problem is to define what "otherwise horrible" exactly means). Actually it is more complicated by this, but you are already missing the very basics.

I know there's a tradeoff here with driving traffic to your Substack

Why not post the contents of the papers directly on Substack? They would only be one click away from here, and would not compete against Substack.

From my perspective, adacemia.edu and Substack are equally respectable (that is, not at all).

request that you be downvoted to but not below zero so you can still post

This would get an automatic downvote from me.

If you get downvoted, write differently, not more of the same plus a disclaimer that you know that this is not what people want but you are going to write more of it regardless. From my perspective, the disclaimer just makes it worse, because you can no longer claim ignorance.

If someone is too wrong, and explicitly refuses to update on feedback, it may be impossible to give them a short condensed argument.

(If someone said that Jesus was a space lizard from another galaxy who came to China 10000 years ago, and then he publicly declared that he doesn't actually care whether God actually exists or not... which specific chapter of the Bible would you recommend him to read to make him understand that he is not a good fit for a Christian web forum? Merely using the "Jesus" keyword is not enough, if everything substantial is different.)

I am not sure how to proceed with my alignment work in the absence of a quorum of people willing to upvote the most valuable and important parts of my research agenda.

I think the answer is: don't.

Going by the feedback, your research agenda seems valuable and important to you, but not to the LessWrong community. So there is no reason why LessWrong should host your Sequence. (I would tell you to put it on Substack instead, but you already did.)

In my neighborhood there are some similar activities.

About once in a year there is a "Money-Free Zone", which means that someone rented a big room for a day (for example a gym at a school, during a weekend) and put there some tables. People who want to donate stuff come there, give the stuff to organizers, the organizers sort it out and place it on the tables. Then everyone is free to take whatever they want. At the end of the day, organizers put the remaining stuff to bags and offer it to some charities, and I suppose whatever is rejected ultimately gets thrown out.

This requires some money and work, but only on the side of the organizers. For everyone else it is free. For people like me it is actually a good opportunity to get rid of some things I no longer need, so I usually give about as much as I take. The event is open for everyone, and giving is purely optional.

The problem is that this works okay as long as giving and taking is at least somewhat balanced. I do not need to take as much as I give, but if I take literally nothing, it removes a large part of the incentive to come the next time. Most of the time it is okay -- I suppose because most poor people do not get the memo? though that explanation sounds a bit weird -- but I have heard that at some places the event was overrun by hordes of poor people (sometimes poor smelly people) which was a bad experience for the donors, so the next year the event was organized at a different location and was not advertised publicly; it was still open for everyone who came, but you needed to be lucky and get the info through the grapevine.

We also have a neighborhood group on Facebook, and related to it there is a mutual donation group, that only the members of the neighborhood group can join. If you want to get rid of something, you post a photo and a description, and the first person who replies can take it.

There is also a website for people selling to each other, where you can also "sell" for a price of 0 €.

Compared to the American versions, as described on Wikipedia, it seems to me that our local version is much less ideological. Like, the Facebook group is not ideological at all, the spirit is "neighbors offering stuff to each other"; at the selling website the spirit is "this is so cheap that I am actually not even asking money for it". Only the Money-Free Zone has some ideological connotations in the title (it may appeal to people who believe that money is a bad idea in general), but the activity itself is very factual: you bring stuff, you take stuff, no one is giving you lectures on anything. I suspect that whatever is your motivation for organizing such events, not pushing your ideology on the participants makes it a better experience.

(I assume that the main effect of a "Buy Nothing Day" is that people buy that stuff on the previous or the next day instead, so the weekly sales remain the same.)

EDIT: This was supposed to be a reply to the answer by Valdes, but for some reason LW keeps posting it as a separate answer. No idea why.

Most people are incompetent, and the competent ones are usually busy. So unless you pay the market price (quite high) or the project is super exciting (to someone other than you), you will get a crappy app. It's not just the code, it can be a crappy design, or utter lack of empathy with the user.

Recently I use an app to record my daily medicine usage. The idea is that every day I take a pill, and I confirm that "yes, today I took a pill at X o'clock". (Or: "today I didn't". Or, I forget to enter either information today, so tomorrow I enter it also for the previous days.) Then it uploads the information to a server. How difficult can this get?

• First I start the application, and I need to enter a password. Then it takes about a minute to authenticate me on a server. I appreciate the concern for privacy, but why the fuck can't the password just be verified locally?
• Then I need to choose whether I want to report the pill usage, or read the tutorial. I have already read the tutorial, why can't it remember this simple fact and skip the screen?
• The next screen tells me that I have reported the pill usage for yesterday, but not for today. Thank you, Captain Obvious, that's like 95% of situations when I use the app. Why can't you just skip this screen in such case, and only display it when something unexpected happens, such as I have already reported the pill usage for today, or I forgot to report it yesterday?
• Then it asks me whether I took the pill today, and I need to check the "yes" or "no" option, and then click Next.
• Then there is a screen that tells me to select time. I need to click a clock icon, it displays a modal dialog where I adjust the hours and minutes (by clicking small "+" and "-" buttons below them; if I click outside the small buttons, the modal dialog closes and I need to click the clock icon and enter the time again). Why couldn't these two screens plus the modal dialog replaced by one screen that displays the hours and minutes with the "+" and "-" buttons, plus another button "I didn't take the pill today"?
• Then there is a screen telling me to review the information I entered, click the "ok" checkbox, and then click Next. Except it doesn't show the entered information, so the only way to review would be to click the "Back" button and check on the previous screen.
• Then it take another minute to upload the information to the server, and then I can finally close the app.

I mean, it could worse, and you can get used to it, but I could also be way more convenient. But it is not, quite predictably. Crap is what you get by default, and there is no market mechanism to select for a higher quality app.

That's planning for failure, Morty. Even dumber than regular planning.

- Rick Sanchez on Mortyjitsu (S02E05 of Rick and Morty)

I imagine an explanation that gets progressively longer. One-paragraph summary. One-screen explanation. One article containing the gist of the Sequences (with links to Read the Sequences), a brief history of the rationalist movement, frequently used concepts (with links to explanations), maybe a list of frequent myths.

Technically, all three could be at the same URL.

The "good" AI (one that cannot output <bad> tokens) would also be unable to explain why someone or something is bad, or warn you about a danger. The advertiser's dream, perhaps, but not really friendly.

Related: use–mention distinction

The rules may be nice, but they are not going to enforce themselves.

Many communist countries had freedom of speech and freedom of religion in their constitutions. But those constitutions were never meant to be taken seriously, they were just PR documents for the naive Western journalists to quote from.

I think it is good to use your goals as a general motivation for going approximately in some direction, but the opposite extreme of obsessing whether every single detail you learn contributes to the goal is premature optimization.

It reminds me of companies where, before you are allowed to spend 1 hour doing something, the entire team first needs to spend 10 hours in various meetings to determine whether that 1 hour would be spent optimally. I would rather spend all that time doing things, even if some of them turn out to be ultimately useless.

Sometimes it's not even obvious in advance which knowledge will turn out to be useful.

When editing the article body, select a block of text, and a menu appears (the one with options like bold and italic). Choose the "f(x)" symbol, with tooltip "Insert math". The syntax is that of TeX.

(\epsilon, \delta) -> $(ϵ,\delta )$

B \subset A -> $B\subset A$

I have no data on OpenAI situation, but #8 has crossed my mind. (It reminded of the communist elections where the Party got 99% approval.) If Sam Altman returns -- and if he is the kind of person some people describe him as -- you do not want to be one of the few who didn't sign the public letter calling for his return. That would be like putting your name on a public short list of people who don't like the boss.

Of course, #5 is also likely. But notice that the entire point of having the board was to prevent the #5 reasoning to rule the company. Which means that ~all OpenAI employees oppose the OpenAI Charter. Which means that Sam Altman won the revolution (by strategically employing/keeping the kind of people who oppose the company Charter) long before the board even noticed that it started.

(I find it amusing that the document that people in communist Czechoslovakia were afraid not to sign publicly, so that they don't lose their jobs, was called... Anticharter.)

The first rule of overcoming ADHD club is: you do not distract me by talking about the overcoming ADHD club.

This is basically math (and computer science) education. On one hand, some parts are probably not very useful. On the other hand, some people expect that teachers will defend every single step along the way by explaining how specifically this tiny atom of knowledge improves the student's future life. No, I am not preparing a PowerPoint presentation on how knowing that addition is associative and commutative will make you rich one day.

It's just saying "the universe's first cause possesses self-awareness and intentionality", which as far as we can tell is impossible to prove or disprove.

I would still assume that someone who believes such weird hypothesis for no good reason probably also believes other things for no good reason.

(If there is no evidence either way, there is still Occam's razor. If you ignore it, so be it, but I will assume that you also ignore it in other situations.)

I would probably put much less trust in their statements containing "it's going to be okay", because I would assume that their assumed reason why things are going to be okay is a supernatural intervention.

And practically zero trust in any excuses they make about history of religion (such as "actually, inquisition was not that bad; they only tortured you if you were a bad guy, and they didn't hurt you much"), because there is practically an entire industry of motivated thinkers whitewashing the history of religion.

But on topics of practical life, there is little difference. Well, except for some sensitive topics such as gender norms and sexual behavior, because I would expect that they talk about what should be (according to Bible) rather than what actually is.

.

By the way, you are probably not going to find many people who believe in God here. If you want answers from a smart but religious audience, you might have better luck at Astral Codex Ten. (For maximum engagement wait until November 27th when Scott makes Open Thread 304, and post there. The current Open Thread already contains too much.)

my failed joke was taken quite a bit as a serious statement

I think you would get actual downvotes in such case (the ones that matter for karma), not these ones.

the recent case of America responding to Russia denouncing the breaking of its promise not to expand NATO eastward with "Ha ha, you should have gotten it in writing!"

I thought the response was more like: "this never happened, you just made that up". You can't break a promise you never made in the first place.

Is the concept of honor so alien that you consider it "incredible" when men show it?

It would be hard to make it work in democracy. A politician may promise something, and get replaced by another politician later. Should the latter honor the promises made by the former? (Is Biden required to build the wall, just because Trump promised it?) The entire point of democratic elections is that people can replace the politicians they no longer want.

This assumes that the costs scale with population size. I would naively assume that it is mostly fixed costs (developing the software, setting up the central servers).

You don't need a fleet of spaceships. One spaceship could throw thousands of nukes from distance (like meteorites) so that they all reach different parts of the planet at the same time. Add thousands of decoys; the defenders will not be able to shoot down all of that.

If the attack is against Earth, you could instead throw thousands of capsules containing highly contagious deadly airborne virus, releasing the virus in the atmosphere.

Or both. Throw the virus first, the nukes two days later. People will bring the deadly virus to nuclear shelters.

I would aim to hire the best teachers (conditioned on a very strong understanding of the material, obviously), while for the graduate school I would aim to hire the best researchers, who would have to teach fewer courses since they would never teach undergrads.

This seems like an obvious solution, so I wonder whether some institutions are already doing it, or there is a catch that we didn't notice.

(This is just a wild guess, but it perhaps a university that only does a half of that -- i.e. hires best teachers and mediocre researchers, or best researchers and mediocre teachers -- would be just as popular, for half the cost. You cannot get unlimited amounts of students anyway, so if you already get those who want the best teaching, you don't need to also attract the ones who want the best research, and vice versa.)

I was thinking from the opposite direction, whether it would make sense for the professors to make pairs -- one who wants to teach, plus one who wants to do research -- and trade: "I will teach your lessons, if you write my thesis and add me as a co-author to your publications". Not sure if this is legal. (Also, it seems fragile: if one decides to quit or gets hit by a bus, the other's career is also over.)

I would imagine "Sort by Controversial" as someone telling the story and having "flashbacks" to what happened. Not sure if that would work ok.

With "Samsara" the point is how the protagonist changes (unknowingly to him) from a normal person into a hardcore spiritual guru (while trying to be the opposite). That should be reflected by gradual changes in how the person dresses, walks, talks, etc. Even the "douchey/slutty" dress code for his followers should be suspiciously uniform-like (e.g. all having exactly the same color). I imagine examples of specific (anti-)spiritual exercises: "imagine a red car... it is a Porsche... very expensive... now focus on the front wheels... inhale... exhale... now focus on the back wheels... inhale... exhale..." said in the hypnotizing voice. (Or maybe with a tone of excitement at the beginning of the movie, but the calm hypnotizing voice later. The point is how the wannabe anti-guru converges to a regular guru anyway.

I imagine that the Blue Eyes might benefit from adding a short visual explanation (a bit like a Khan Academy video) for how exactly e.g. three people would in three days figure out that they have blue eyes. What I am trying to say is that sometimes a picture explains an idea better than a dramatic dialog.

OK, looking forward to your videos!

Most essays in the Sequences are not suitable for epic movies, but there are the "beisutsukai" stories:

• Initiation Ceremony
• The Failures of Eld Science
• Class Project
• The Ritual
• Final Words

Some stories from Slate Star Codex / Astral Codex Ten:

• It Was You Who Made My Blue Eyes Blue
• Samsara
• Sort By Controversial
• …And I Show You How Deep The Rabbit Hole Goes

I think I would like to see Samsara, that would be fun!

My question is who this system leaves better off? Students get worse professors, good researchers have to waste their time teaching and good teachers have to waste their time researching.

I am quite curious about this, too.

I suspect there might be some kind of fallacy involved, something like "if we make a job that is for both research and teaching, we will automatically get people who are good at both research and teaching... even if we actually evaluate and reward them only for the research". Maybe, if someone sucks at teaching, it is assumed that they would never apply for such job in the first place -- they could get a job at some purely research institution instead. (So why does this not happen? I suppose that even for a researcher without teaching skills, a work at university can be preferable for some selfish reasons. Or they can be overconfident about their teaching skills.)

And the following step is that someone who is good at both research and teaching is obviously better than someone who is merely good at teaching, because such person will be able to teach the latest science. Which ignores the fact that a lot of what is taught at universities is not the latest science. But it is still better to have someone who has the ability to get the latest science right.

To steelman this position, imagine the opposite extreme: imagine a university where all teachers are great at teaching, but suck at research. It would be a pleasant experience for the students, but I would worry that a few decades later what the professors teach could be obsolete, or even outright pseudoscience. Also, teachers who are not themselves good researchers might have a problem to bring up a new generation of researchers; and where else would we get them?

some professors feel like once they have tenure they don't need to pay attention to how well they teach.

I imagine that if they taught well before, they would still teach well by the sheer force of habit. Maybe slightly worse because they no longer bother to do it perfectly, but not "consistently present things in unclear or inconsistent ways".

The main alternative is to be a consumer, who simply spend it all.

Or have it taken away, and given to someone who is better at spending.

“How do we make AI benevolent?” is a badly formulated problem. In its very asking, it ascribes agency to the AI that we don’t have to give it

Yes, we don't have to, but considering that people are already trying to give agency to GPT (by calling it in a loop, telling it to prepare plans for its future calls), someone will do this, unless we actively try to prevent it.

As someone whose job it is to examine and improve the structure of agency and clarify values, I can say with confidence that as a culture we have only a very primitive understanding of either. 

100% agree. But that's exactly the point. MIRI is trying to solve alignment not because they believe it is easy, but because they believe it is hard so someone better start working on it as soon as possible.

The hope is that “we” (meaning, someone) can somehow tell AI the final answer about what we should want, or get it to tell us the final answer about what we should want, and then leave it to execute on our behalf all of the weighty decisions we are not competent to make ourselves. We should be very wary of a project to save ourselves, or even “empower” ourselves, that is premised on the belief that humans essentially suck.

I read the news about the war in Ukraine, or Israel and Palestine, and it seems to me that humans suck. Not all of them, of course, but the remaining ones suck at coordination; either way, the results are often bad.

The final answer we tell AI could include things like "take care of X, but leave us free to decide Y". Maybe, don't let people murder or torture each other, but otherwise let them do whatever they wish? (But even to achieve this seemingly moderate goal, the AI needs to have more power than the humans or group of humans who would prefer to murder or torture others.)

Yes, there is a risk that instead of this laissez-faire approach, someone will instead tell AI to implement some bureaucratic rules that will strangle all human freedom and progress, essentially freezing us in the situation we have today, or perhaps someone's idea of an utopian society (that is dystopian from everyone else's perspective). However, if such thing is technically possible -- then exactly the same outcome can happen as a result of someone acting unilaterally in a world where everyone else decided not to use AI this way.

Again, it seems to me like the proposal is "there is a button that will change the world, so we should not press it", which is nice, but it ignores the part that the button is still there, and more people are getting access to it.

It would be better to de-fixate on the arms race, and instead imagine applications that are built to help ground people in reality, to explore where and why they respond to which sensations and drives, to know themselves better and give themselves more grace. 

I 100% agree with the idea of using AI for self-improvement.

A practical problem I have with this is knowing that current "AI therapists" have zero confidentiality and report everything to their corporate masters, who will probably try using this knowledge to increase their profits. They will probably also try to increase their profits by nudging the "AI therapist" to give me certain ideas or avoid giving me certain ideas. Thus a Microsoft-sponsored therapist might tell me that Linux is a waste of time, and explain how not trusting our corporate overlords is just a part of teenage rebellion that I should already be mature enough to overcome; a Meta-sponsored therapist will encourage me to develop more contacts with people using social networks; and a Google-sponsored therapist will encourage me to buy whatever the highest bidder wants me to buy. The information they get about my weaknesses and worries will be leveraged to do this more effectively, while explaining to me that my lack of trust is just a childhood trauma I need to overcome.

But, ignoring this part, if I could believe that the AI is impartial and keeps our discussions confidential, of course I would use it, among other things, as a therapist and a self-help coach.

But even if 99% of people use it this way, it does not remove the problem of what if the existing dictators and wannabe dictators use it to increase their power instead, automating whatever they can.

I really hope this has nothing to do with Sama's sister

That reminds me of the post we had here a month ago. When I asked how exactly are we supposed to figure out the truth about something that happened in private many years ago, I was told that:

• OP is conducting a research; we should wait for the conclusions (should I keep holding my breath?)
• we should wait whether other victims come forward, and update accordingly (did we actually?)

Now I wonder whether Less Wrong was used as a part of a character-assassination campaign designed to make people less likely to defend Sam Altman in case of a company takeover. And we happily played along.

(This is unrelated to whether firing Sam Altman was good or bad from the perspective of AI safety.)

Sorry if I missed your preference stated somewhere, but what would be your position on linking some of your new articles from Less Wrong in the future?

I'm in the process of gathering more information ... For now, I'll suggest that you wait until I finish up my research

So, is there any progress on this topic?

So, to put it shortly, there are more of them, but most of them are not what you think.

This feels a bit confusing. I mean, if they are not what I think they are, could it mean that we are kinda talking about two different things? There are many "people on autistic spectrum". There are few "rain men". It's not that the former statement is more true than the latter; they are statements about different things. Are we actually debating the territory here, or just disagreeing about what the label "autist" should refer to?

(Is the conclusion I am supposed to take here something like: Many people are pedophiles, but most pedophiles actually don't want to have sex with kids... they just think that kids are kinda cute? Many people are psychopaths, but most psychopaths do not lack empathy... they just disagree with some effective altruist ideas?)

In case of autism, the statement about territory is that sometimes the things you assumed to be binary or at least bimodal are actually on a continuum ranging from normal to the extreme. And therefore... you should not have a word for "people at the extreme end of this continuum"? Words exist for a reason; apparently someone finds it useful to talk about the extreme end. Perhaps we should just be very careful to avoid a fallacy of assuming "because we have a special word for it, it must be binary or bimodal". (And yet somehow, we can talk about "tall people" without assuming that height is bimodal. Perhaps because height is not a sensitive topic?)

But this is probably not what you wanted to say. The lesson about homosexuals was supposed to be "there are more homosexuals than you (an openly homophobic person, or a person living in a homophobic society) assume", not just "sexual orientation is a continuum"; even if the latter is also true. Similarly, the lesson about sexual assault is not just "sexual assaults happen on a continuum... from rape at one extreme, to just saying hello or smiling at someone who doesn't want to be smiled at". The lesson is that there are actually more sexual assaults (in the usual meaning of the word) than we assumed (but coming from people who often do not match our expectations, leaving victims who often do not match out expectations).

...in other words, I think you are talking about an important and true topic, but when I try to put it in exact words, I find the concept slippery. Could it perhaps be two or three different things? ("a typical X is not what you imagine" vs "there are more X than you imagine, even using your current idea of X") Or both at the same time? ("there are more X than you imagine, and even more of what you would call partially-X or X-lite")

The young don’t even know how to use computers anymore.

Did they ever?

I mean, this could be a typical-mind fallacy. Many of us (LessWrong readers) knew how to use computers as kids, but most kids of the same age didn't. It could be the same today.

Thank you both! This is perfect. It's like a rational version of Twitter, and I didn't expect to use those words in the same sentence.