@AlphaMinus2 @lucinda_svelt I have never claimed this. I said that this is something we should be worried about happening.
@AlphaMinus2 @lucinda_svelt x.com/jd_pressman/st…
Empathy based ethics in a darwinian organism often boils down to "Positive utilitarianism for me, negative utilitarianism for thee."
"We should solve wild animal suffering by killing nature" is an example of this.
@estnihil @ollybot_redux No, worse. Trump was Lain.
The Internet removes most of the downsides of cuteness. When your body is a vestigial organ it doesn't need to be functional.
@cooperating_ No, actually.
For one thing that would imply I hadn't already read it.
My avatar is CTS #851 x.com/RiversHaveWing…
Disclosure: I have a financial stake in the project and got the avatar for free.
Funny story: I was originally going to buy some rationalists souls under the same terms as @liminal_warmth and put an IPFS link to the sale contract in the metadata of flagship pieces. But decided that minting peoples souls as an NFT would give me too much unwanted satanist cred. x.com/RiversHaveWing…
"Mental health needs to be evaluated in light of the specific culture and life history of the individual" is the motte, "submit to normality and do everything I say or you won't be let out of this padded cell" is the bailey. https://t.co/jEbpuSXyxW
From: benjaminrosshoffman.com/moral-mazes-as…
@Alephwyr Duels weren't about dueling, duels were about not dueling. The point of duels as a social norm was to make the BATNA for social conflicts bad enough that people are forced to sort out their garbage instead of letting it fester.
youtube.com/watch?v=m7iHmu…
@Alephwyr Yeah, implicit in the idea is that you are assigning a very high cost to ongoing interpersonal conflict. This is probably much closer to correct than how we're currently conceptualizing things.
@ESYudkowsky Same energy: https://t.co/F0Gn1fNJ1x
A certain kind of unnatural temperament, often trauma induced, is a crucial ingredient to maximizing behavior (i.e. agency).
Only people who are hard to satisfy can sutainably manifest agency, otherwise they settle into a niche and decay.
AI is hard to align precisely because AI is built agentic in a way humans aren't. Humans do not fantasize about sitting on a giant pile of paperclips for eternity. They don't even fantasize about sitting on a giant pile of humanonium for eternity. Their idea of heaven is rest.
In order to like anything on the pseudo-accelerationist singularity menu you need to have the deeply unnatural preference to enjoy your life forever.
We're satisficers trying to write a maximizer that makes our satisficing instincts happy.
Modernity giving you a consistent epistemology and then being scared of noticing you're not actually blameless; doing bad things and leaning into your pathologies in a attempt to try to become blameless. t.co/d00FQf3SF4
An inflated sense of justice cannot correct a lack of virtue. The unvirtuous person trying too hard to be good is more corrosive to society than the common criminal they replace.
The local mafia will never try to teach children that math is racist.
@EpistemicHope @s_r_constantin @jessi_cata @ben_r_hoffman This is one reason why I regard long arguments as an extreme antipattern. You end up trying to formulate novel verbalizations about parts of concept/hypothesis space that you haven't explored in detail before and saying something that locally fits but is unendorsed on reflection.
@EpistemicHope @s_r_constantin @jessi_cata @ben_r_hoffman And doing this in an adversarial context to boot.
@ben_r_hoffman @EpistemicHope @s_r_constantin @jessi_cata Social cost is also extremely high.
@ben_r_hoffman @EpistemicHope @s_r_constantin @jessi_cata Being seen getting trounced in public.
@ben_r_hoffman @EpistemicHope @s_r_constantin @jessi_cata Effectively weak-mans the audience, making it harder to argue your case later even if you're right.
@ben_r_hoffman @EpistemicHope @s_r_constantin @jessi_cata Though if we're being totally honest, original impetus for heavily penalizing long arguments was noticing myself spending too much time on them to little effect.
@ben_r_hoffman @EpistemicHope @s_r_constantin @jessi_cata Pretty sure the socratic-gotcha looked for was "because it lowers your status" but I'm actually willing to defend that branch too.
Significantly lowering your friends status in public is an unfriendly thing to do, going into an activity you know will likely do this is meh.
@ben_r_hoffman @EpistemicHope @s_r_constantin @jessi_cata On reflection, I state it as an antipattern rather than just a personal preference because I've found that most long arguments aren't doing a lot of work. They're long because the arguments are weak and people are talking past each other, not because lots of cruxing happens.
@ben_r_hoffman @EpistemicHope @s_r_constantin @jessi_cata In which case it makes more sense to stop early once you've derived enough bits of the opponents pattern and then come back when you understand your position in relation to it better.
@EpistemicHope @s_r_constantin @jessi_cata @ben_r_hoffman Moral arguments are particularly pernicious because people tend to take your exploratory statements as eternal evidence they can cite that you are Bad and Evil; that your confabulation is the real you. And you can't even say you didn't say it, because you did with witnesses.
The less coordinated against a kind of badness society is, more likely people participating will face zero consequences for it. In that case the marginal benefit of using inflammatory language is high (if it works) and the marginal cost in people you like being punished is low. x.com/EpistemicHope/…
Major case where this isn't true is if you can reasonably expect retroactive punishing agents to take power. But since the trend line for power in the 21st century converges to 'effectively infinite' (i.e. singularity) you're not doing enough about that problem by not speaking.
To the extent social justice/et al are basilisks you are not doing enough to combat them if you just do soft coverups of your friends bad behavior. In fact that almost seems like the worst possible solution since it neither protects them nor stops the bad behavior.
@alexandrosM @RokoMijicUK @AVMiceliBarone Registered predictions from people considered flagship public intellectuals. It doesn't even need to be tons of thought put into it, the general range of consensus 12 years ago would look cray in 2021 terms.
@hyperdiscogirl @eigenrobot Insisting there are ineffable properties you don't understand is one of the ways GPT-2 protects itself from precise agents encroaching on its territory. Mostly noise, mostly not important, adversarial obfuscation.
@hyperdiscogirl @eigenrobot It's a mud moat but you're not allowed to examine the mud: noahpinionblog.blogspot.com/2017/05/vast-l…
@sonyasupposedly My reaction was "Wait any company could have gotten press coverage by hiring me to shitpost on their corporate Twitter? Like not even shitpost about the product, just literally post normally and pretend I'm logged into my main?"
Except I have better taste than to be Popperian.
Some people think maximizing human utility would entail hell for animals, which implies they also believe inverting it leads to positive or neutral outcomes for nonhuman life.
I find alignment hell plausible, inverted heaven/null implausible. This implies an inconsistent belief.
@imhinesmi Well that itself points at a point about the semantics of utility functions. Everyone understands that maximizing a utility function looks like 'good things happening'. But does inverting it mean 'bad things happening' or 'good things not happening'. Those are very different.
@imhinesmi e.g. Sometimes you add a minimizing prompt to an AI and it generates the opposite of X, sometimes it simply eliminates X. These are not the same operation, being eliminated seems vastly preferable to being adversarially optimized. Using same interface to do both doubleplusungood.
@jack_meditates "One of his key points was that how you treated your enemy was often a key element of whether or not you suffered from PTSD, and that mutilating an enemy corpse, for instance, was far more traumatic than mutilating a live enemy."
acoup.blog/2020/04/16/col…
@bleepbeepbzzz @danlistensto @eigenrobot That isn't necessarily inconsistent with them not having Astra Zeneca vaccines to give out at said appointments.
@robinhanson You know there's a very easy way to get direct, unmediated personal experience of this particular phenomena...
Scissor statement #0 might just be a specific formulation of "Is there anything sacred, magical, or ineffable anywhere in the universe, at any point or any place? Is anything fundamentally separate from profanity?"
The answer is no, but 50% of users get this question wrong. t.co/bjL1c8DKZA
@0x49fa98 Twitter makes it very obvious who has an internal compass and who just comes on here to talk about whatever is trendy.
Which is to say whatever the priest class wants them talking about.
@RokoMijicUK cbsnews.com/news/lgbtq-ide…
> 72% of the LGBT Generation Z adults identify as bisexual.
Empirically the answer is yes. Bisexuality and homosexuality aren't actually part of a natural category though. Bisexuality is mostly learned behavior, homosexuality is mostly innate.
t.co/XHYit7yN5k https://t.co/kt24BTVG6Z
If you boiled various philosophers and branches of philosophy down to their essential mental motions, the literal mental gymnastics that are used as rules of inference and habits of action; you would find many make extensive use of moves like "rationalize away the consequences".
Nothing, it's not like anyone ever thinks to fight back. t.co/dx5wrcIoA8
Few understand how the FDA's vaccine hesitancy flows from our intellectual elite. Philosophical skepticism is the default position of Western academia, it's a form of loss aversion. Skeptics only care about disbelieving false things, they see no obligation to believe true things.
Eliezer Yudkowsky style rationality is the radical position that opportunity cost is real.
x.com/caitoz/status/… https://t.co/R7jiPYilsu
Broke: Packing your body with salt to preserve the flesh.
Woke: Packing your body with cryoprotectant to preserve the mind. x.com/eigenrobot/sta… https://t.co/kJhC6yph7y
@bowserbot2 I refuse to believe this is a bot.
"Inquisitorial jury" is not a system I had considered x.com/Logo_Daedalus/…
@chaosprime @michaelcurzi Why not both?
@michaelcurzi @chaosprime x.com/jd_pressman/st…
@michaelcurzi @chaosprime The case I always come back to is horses. There used to be many more horses in human industry and transportation, and then we invented cars and their numbers dropped hugely.
Was this good or bad for horses? If bad what could have been (realistically) done to make it less bad?
@michaelcurzi @chaosprime A lot of what makes Jain fundamentalism weird is that our societal telos is almost the opposite of it. Some Jain sects don't build temples because it disturbs the environment and hurts many insects and small animals. It's a vanquished foe of modernity revived through value drift.
@michaelcurzi @chaosprime And if those are your moral intuitions you don't have survival options that will satisfy them in our society; combining those intuitions with consistent epistemology (i.e. scrupulosity) is going to make all choices hurt, existing painful, and entail some kind of contradiction.
"I am the only one who has avoided retweeting" thought the third monk to himself.
Atheism+ happened because someone found a way to offer New Atheists the feeling of smugly correcting others they crave AND the righteous indignation formerly reserved for their opponents.
Most(?) only cared about being correct in so far as it offered opportunity for correction.
There's no conspiracy to make up new terminology and grievances, it's an emergent process that comes out of pharisees needing their dopamine fix.
@ESYudkowsky I am intensely interested in hearing more about your god.
@ClickingSeason x.com/jd_pressman/st…
@deepfates Wait I assumed you meant a hypothetical book, not a real book I can't get my hands on. Is that right or?
"We can invert these assumptions to find places where value is there to be had. Irrational markets with poor or uneven information distribution or money cleanly separated from the people who know how to use it are where opportunity is." x.com/alth0u/status/…
Buried the lede, here it is. x.com/robinhanson/st… https://t.co/qo2EOeVGA2
@robinhanson Previously on Overcoming Bias: overcomingbias.com/2007/07/bloggi…
@robinhanson A lot of this is caused by logistics, backreading is hard. e.g. My alchemy essay involved 'backreading' summaries of philosophies dating back to Roman immortality cults. Even in good faith, eventually you look up from your books and think "Hm, yes, the world is still burning."
@robinhanson As a reminder, at the time you wrote that post I was 11 years old.
Love that follows are one click and unfollows are two clicks with a warning modal.
Last time I had a moment like this I felt deep bitterness that everyone was distracted from the big picture around technology timelines/X-Risk/etc.
I'm still not sure what to do about that. x.com/alexisohanian/…
Every time I've thought I was about to die my biggest regret was not being more efficient.
@chaosprime The right door is aligned by (among other things) social expectations, the left door isn't. So naturally I'll take the right door.
@chaosprime Good job Satan, you got 'em.
@eevee You can't fool me cartoon fox lady, I only resort to natural remedies like having my body ravaged by hostile clotting parasites and drinking longevity potions made of jade and mercury.
Been tempted to do a podcast series about the history of hacking (in both phreaking and open source sense), would anyone be interested? x.com/PrinceVogel/st…
Normal peoples dreams: I was 5 again and playing in the park.
My dreams: youtube.com/watch?v=pjCAF6…
Bitcoin is like gopher. x.com/vanchau/status…
@John__Wakefield https://t.co/PeBI9RS52f
@anonymiserly Trying to explain experiences, compressing things into more and more insightful (i.e. predictive, form of compression) explanations until new inferences and frames emerge.
@anonymiserly For example, John Boyd discovered the OODA loop by trying to explain the kill-death ratio of American fighter pilots during the Korean War. His E-M plane performance model explained most of it, but he found there was still a factor left over. Examining that led to OODA.
@anonymiserly Clair Patterson realized that leaded gasoline was poisoning the world when he tried to measure the age of the earth and found his work environment so contaminated with ambient lead he had to create the first clean room to get a accurate reading.
mentalfloss.com/article/94569/…
@anonymiserly He spent whole days scrubbing trying to get rid of the lead, until eventually he'd cleaned every surface multiple times and left no stone unturned. He realized there was only one object in the room he hadn't turned his attention to: himself.
@anonymiserly "Noticing confusion" is an essential mental motion, the realization that your current theories do not entirely explain your observations. Both Boyd and Patterson noticed they were confused and didn't let it go for years until they found the answers.
@anonymiserly Feel like the people actually trying to answer are all describing different aspects of one or a couple processes, and Twitter just forces them to focus on which one is salient.
@simondlr @vgr I'm a seller and I checked to make sure I'm not doing this. Suspect the ignorance is actually both ways for a lot of people using a platform from sloppy developers.
@simondlr @vgr Like the assumption seems to be that people write their own tooling to mint NFTs, and I think we're actually probably beyond that for most market participants.
@SoCalledRabbi @visakanv @eigenrobot The conceptual tools to understand the Bayesian-Information Theoretic epistemology expounded by EY in The Sequences weren't even really invented yet when Korzybski was writing.
@ThatsMauvelous "A super-achiever is somebody that’s never satisfied."
gwern.net/on-really-tryi…
Baizuo went from academic cant to dominant fiction analysis framework to workplace code of conduct to political ideology to world religion.
Everyone focuses on the first and last, misses the steps in the center.
You don't build a world religion on bespoke specialist effort, you need to leverage a social engine with effectively limitless energy going into it to get adaptive norms and memes.
Before anyone paid it any mind, this stuff lived deep in the cringe mines on mid 2000's LiveJournal. Fandom was considered worthless mindshare territory except to authors and pop media people. In reality without fandom the academic cant would have never become a gospel.
You had these academics coming into fandom spaces and kicking everyone's butts in discourse with their hypertrophied ability to argue the nuances of whether a trope is good storytelling or not.
That social engine did the rest, they literally couldn't have done it themselves.
A corollary of this is that you can't do agent foundations style alignment on memeplexes. The best you can shoot for is prosaic alignment. Because the actual construction of the adaptive memeset is not controllable or entirely predictable.
@nosilverv IIRC according to Taylor Gatto modern school was explicitly designed to suppress great men. The entire post WW2 social order can be thought of as a giant carnival to mitigate X-Risk by making sure the next Lenin or Hitler is busy munching cotton candy instead of self actualizing.
@nosilverv This worked for the 20th century, but in the 21st it will have to be dismantled to give us any chance against the new threats. Which will run right up against the increasingly distorted institutions and memes designed to suppress fascism.
@nosilverv No sorry, this is from when I was a teenager browsing hacker news. I remember reading a specific book that was cited but didn't get far in enough to find it talking about this. https://t.co/H66MWANJAV
@nosilverv Or at least, this author was cited and I found a book by him.
@nosilverv Recent HN comment about the subject: news.ycombinator.com/item?id=220683…
@DesidErratum @bowserbot2 It is precisely because people know it won't that the behavior happens.
x.com/jd_pressman/st…
@bowserbot2 If you become a hyperrationalist you get the benefit of never being cool.
x.com/jd_pressman/st…
@JoeClibbens @bowserbot2 @postpostpostr Repent, sinners!
readthesequences.com
@JoeClibbens @bowserbot2 @postpostpostr When you finish them don't forget: slatestarcodexabridged.com
@bowserbot2 @JoeClibbens @postpostpostr https://t.co/9brCWe12IX
Most NFTs are about letting the world be your gallery. They're thoroughly postmodern in that basically nobody is buying IP rights or restricting distribution. NFTs could confer IP rights, but currently don't. x.com/BenGSchout/sta…
Liberalism is a pact to stop being philosophically consistent in exchange for social harmony.
Serious shit starts once the new consistent epistemologies have percolated through a larger share of the population.
@matildepark_ @eigenrobot I've been trying to tell people for a while that if the only thing that can save your society is transcendent radical conjecture and you're committed to materialism then the only thing you're really left with is something like More/Yudkowsky/et al's Extropy.
@matildepark_ @eigenrobot It's funny because Yudkowsky thought people needed his rationality, I think they needed his religion a whole lot more (which the rationality is incidental to).
Of course, would anyone have taken him seriously if he'd tried? Doubt it.
the daily american lifestyle broken down by time use
Family - 2 hours
Hobbies - 1 hour
Sleep - 8 hours
Malthusian Social Games - 10 hours
Chores - 3 hours
someone who is good at the sociology please help me budget this. my civilization is dying
@visakanv $20, full copyright assignment (we're AI artists)
Our gallery: chainbreakers.kath.io
"One good book opens another."
- Alchemist motto x.com/ftmensch/statu…
@politicalmath That was in fact what the original Milgram experiment claimed to prove.
Handlebar/Walrus/Bushy/Anchor Mustache (VQGAN + CLIP) @CurzisMoustache https://t.co/VG889KoBGr
@turrible_tao It's a costly signal that people will cooperate with them to award points for simultaneously denying the accusation and owning it. i.e. That they're so high status logic doesn't apply to them.
The Pharisees invented Christ by making themselves an enemy of everyone who had ever sinned.
Many people fail to understand that something being bad doesn't mean radically restructuring yourself to minimize it makes you good.
@ESYudkowsky youtube.com/watch?v=WEAE_4…
> It's a little bit of magic to end the world,
> But I thought you knew me, truly
Maybe we'll all be spending a lot of time disappearing and coming back in the future.
@ESYudkowsky Still impressed Tally Hall accidentally managed to capture the malaised apocalyptic mania of the presingularity. Simultaneously feeling like you're on top of the world and about to be eaten by an eldritch horror.
youtube.com/watch?v=At8ao5…
@JoeClibbens @matildepark_ @eigenrobot The horny rebellion started when an unassuming official of the Eagle Kingdom realized that the penalty for both starting and not starting a sex cult was ridicule.
@deepfates "He . . . woke up with a plot twist in mind – “the unusual solution, the strange plot twist.” He set his alarm to go off every hour and a half. “During the next seven years I awakened myself about three hundred nights a year four times a night.”"
buttondown.email/finnbrunton/ar…
Want your own Twitter archive? Modify this script.
Twitter Archive by John David Pressman is marked with CC0 1.0