I will never run out of ideas for blog posts because blogging generates ideas faster than I can blog. The number of ideas for blog posts I have massively outnumbers the number of blog posts I have written. To illustrate my point, here are some of the ideas I have for blog posts.

The list is written in reverse order. Newer ideas are at the top.

List of Blog Ideas

  • Writing is anti-inductive

  • Farming long-tailed upside

  • Re: Misinformation About Misinformation

  • Modern straps

  • Re: "Instead of competing with everyone else for the attention of the median person, it's often worthwhile to make something just for smart people. They may be a small fraction of the population, but there are still a lot of them in total."

  • Book Review of MCTB

  • Study notes from Molecular Biology of the Cell

  • Finish translating The Art of War and the other 6 military classics

  • "Material Construction", a series of posts about science that starts by explaining how atoms work and then builds everything out of that

  • The Virtue of Trolling

  • How Modern Art Works

  • Aluminum fuel cells

  • Evaporative Cooling in AI Alignment

  • Predictive Processing and Placebos

  • Japanese stuff in Star Wars

  • Biting Fingernails

  • Geoarbitrage

  • Dilute your Kool Aid with more Kool Aid

  • Re: PG's How to Lose Time and Money

  • managing client expectations

  • Wave harmonics and brain synchronization

  • Rationality is someone who cares more about being right than feeling right

  • Against the intrinsic value of money

  • What would fractional reserve banking look to a predictive processor?

  • Meeting writers IRL

  • Learning the Others' language よろしくお願いします

  • Advice for Smart People with Autism

  • How to get an internship

  • Re: this comment by Razied about meditation

  • Re: Mandatory Secret Identities

  • Re: Going from California with an Aching in My Heart.

  • Star Wars and Marvel doing same thing with expanded universe

  • Trump symbiosis

  • Interview with Bryan Caplan

  • Everyone should be like us

  • Russia showing torrents in theaters

  • Adult Responsibility

  • Collective Guilt

  • Re: Selfish Reasons to Have More Kids

  • Money is commoditized coercion

  • Proper Repro Steps

  • Absolute reference points. Talking about what people are talking about is a trap.

  • a series on Rationality

  • Minimum Viable Skynet

  • Unbelievable stories go into fiction

  • Busywork

  • Paperclip Maximizer Minimizers

  • Pedagogical techniques from different domains

  • Noninterference with Primitive Minds. Let them figure out your own values.

  • The Wisdom of my Lyft Driver. "What's something you think I should know?"

  • The 4-Hour Cult Leader

  • Buzzwords. Unnecessary when there's universally-understood word instead. But shows you know the new stuff. Illustrate in Effective Evil.

  • Post-Post Modernism. Post-Post-Post-Modernism. End of philosophy. We have solved all the big problems.

  • Copying code from StackOverflow. Use as metaphor for rationalist debate failure modes. Also alien technology imported from China. Link to Americans Are Like Space Aliens post.

  • Daoist Zero-think

  • Just copy my conversation with my squire. Task switching cost of autism.

  • The confusion cycle to learning philosophy

  • Physicists solve problems by figuring out the real answer. Not by pointing out where others are wrong.

  • Challenging kids with "I don't know if you can do it."

  • Stupid model of world model, agent, biases and akrasia. Satire.

  • Instantly hitting it off or not

  • The antidote to cults is to dilute your mind with all kinds of garbage

  • Indoctrinating against indoctrination by indoctrinating into everything

  • Re: How many heresies do you have?

  • Perfect fit or not at all

  • The value of being in a little bit of a rush

  • Cultivating impatience. Life is so, ridiculously, short. Start becoming people at fourteen-fifteen. Adults at eighteen. Three years.

  • Mental health model: dials on an analog computer

  • Delegate only order at a time

  • What would I want Rationalist MeetUps to look like. Something awesome and ridiculous.

  • Fun writing "What if Moses was a rationalist and Ramesses was a Maxist transhumanist?"

  • Post-pandemic business opportunities. Board games. Coffee shops. Restaurants.

  • Better educational media. Video and blogs. Recorded lectures are boring.

  • Re: Life is Short

  • Re: nation-states House-elves

  • Removing obstacles, especially time commitments

  • The exchange value of money. Government will kick in your door.

  • Jobs where you pay someone to use software. Connor's technician job. The Vanguard people I talk to.

  • The high price of on-demand service

  • Rejection research. Tim Ferris' 4-Hour workweek was turned down 26-29 times.

  • My Policy on News

  • Leadership Starts with Commitment. Making events happen.

  • Handicapping to be around less smart people. Do it in a foreign language or when I'm overwhelmed.

  • Feelings on Christianity

  • The right tools Cheatsheet

  • Smart, Proactive, Independent-Minded

  • Legitimacy and production value. The key to looking legit is high production values.

  • Where to find smart people. Where to find independent-minded people.

  • Accidentally stealing from stuff I read before

  • Don't focus on what is wrong. Memecoin discussion.

  • What is a mental disorder?

  • World is too complicated. When I was in high school, I felt the world wasn't complicated enough.

  • Fun is more important than good

  • Comic where weakling holding iron weapon claims muscly barbarian wielding bronze won't stand a chance.

  • What do I like?

  • Thief visits lockpicking club

  • Don't Argue with Idiots

  • "Some people create or discover new things. Some enforce social norms. There is little overlap between the two."

  • Low expectations

  • Software has been writing software for decades.

  • Bad advice I got when starting programming: bad textbook recommendations, especially design patterns

  • Falsifiability

  • The future of combined arms. Offense used to be tank-centric. Now what?

  • The power of low expectations. Morale.

  • Digging worthless ditches

  • YouTube genres and freely-available visuals.

  • Failing on act transitions. The Maze Runner failed in Act 3. My Dress-Up Darling went from "best anime of the year" to merely "good" on the transition from Act 1 to Act 2. Overlord?

  • Using college for job application discriminates against Asian Americans because of affirmative action

  • Comical geopolitics story about Middle Eastern dictator acquiring nuclear weapons

  • Thoughts on Eurocentrism

  • Don't blame the media. In a free ecosystem, you can pick what to pay attention to.

  • The Situation in Taiwan. Causus belli.

  • Building capital. Negative maintenance. Opposite of a job.

  • What it's like to have high conscientiousness

  • The magic of low expectations

  • Optimizing for looking stupid. Looking poor.

  • Location-based pay

  • Meeting more people

  • My Fantasy Human Resources Department

  • How I would run a software company.

  • Why are programmers paid so much?

  • The best job assignment is "same as before" because no negotiation

  • College alternative advice

  • Being awesome instead of learning to be awesome.

  • Could do another Ukraine situation report, this time from a broader perspective. "Ukraine Situation Report - Geopolitics Edition"

  • Visual hierarchy

  • Breaking the Box, from CGI

  • Developing a Visual Library

  • Pareto Hours

  • Constraints give your life shape

  • "Most people would say, I'd take that problem. Give me a million dollars and I'll figure out what to do. But it's harder than it looks. Constraints give your life shape. Remove them and most people have no idea what to do: look at what happens to those who win lotteries or inherit money. Much as everyone thinks they want financial security, the happiest people are not those who have it, but those who like what they do. So a plan that promises freedom at the expense of knowing what to do with it may not be as good as it seems."

  • Everything is useful in non-artificial settings

  • Samsara

  • Everything I can do is boring and easy. Everything I can't do is incredible and magical.

  • Scaling down ambitions.

  • A sequence on material reductionism

  • A sequence on how math works

  • Romance protagonists

  • "Attention Is All You Need" article

  • "There is such a huge disconnect between employers, who think job interviews are merely chatting with the candidates, and the experience of many candidates, who have to prepare for weeks for what is in fact a highly demanding and artificial process."

  • "When people are naturally good at something, they tend to be puzzled by how bad other people are at it. So that's one way to figure out what you're naturally good at. What do other people seem puzzlingly bad at? (Being right doesn't count. Everyone thinks they're right.)"

  • "I am curious what would happen if you pointed out contradictions. For example, if people think 40% of Americans are black and 59% are white and 24% are Asian, and that’s 133%. Which is a lot." Pie charts.

  • Organizations cannot distinguish "truth" from "what I believe to be true".

  • Limiting factor mental exhaustion, not time

  • Highly-conscientious people stressing out for no reason

  • Authentic American magic

  • Theses should be easy to refute, if wrong

  • The "being male" confidence bias

  • The correct refutation is "you're wrong, the truth is " to almost anything. There is one exception, which is "you're bullshit and not worth refuting", which is for hostiles.

  • The quoting trick

  • C'est pire qu'un crime, c'est une faute

  • Authentic American Magic Tricks

  • Geoarbitrage. Tourism and exports.

  • Bringing back the Mainframe

  • Whataboultruism

  • Why not donate everything argument? Because earning more money takes priority over spending.

  • Stuff that's going right in the world.

  • Fermi calculations (chained multiplication) and how uncertainty increases tail risks

  • Sanctions

  • Zvi's Ukraine War posts have almost no intersection with mine

  • Blog vs YouTube debate

  • Units of caring

  • Trying too hard to do good. Premature Altruism. Premature optimization in altruism. Helping out Russian defectors. Two things: negotiating a business deal to apply tech I invented to helping people with behavioral disorders. Then I set a message to a writer who's making a training dataset for an organization that's trying to save the world from killer AI. I wrote an email to a young man I'm helping put his life in order, and last weekend I taught system administration to a teenager. I cooked myself a simple vegetarian meal.

  • Negative motivation and positive motivation in EA

  • Effective Altruism YouTube

  • Nuclear Winter

  • How long until billionaires can go to the moon?

  • US-Centrism

  • Person says . Other person says . Solution: What is true. Then stop paying attention.

  • Japan and Sanctions

  • 2004 Russian invasion of Chechnya

  • "Curiosity seeks to annihilate itself." So does pain.

  • Why Facebook & Discord popular? Opt-in. Random people don't reply to your posts. Less negativity.

  • People I interact with are extremely long-tailed in their impact.

  • Positive Statements Trump Negative Statements. More work to refute a negative than to claim one? Something going on there. Use Lord Voldemort as example. Tyler Durden too. Difference between writing a novel and explaining why the novel is bad. Not worth refuting. Gish Gallops. It takes less effort to generate bullshit than to refute it. Write Timelessly. Understanding. The idea the world should be a certain way.

  • Positivity, negativity and propagandistic Gish Gallops

  • How do small radar cross-sections work?

  • I don't think about probabilities of outcomes because I instead think about strategy. What would I do. Instrumental.

  • Economic Encapsulation

  • Re: How the Weak can Win a Protracted War

  • How railguns work

  • Anti-Inductive Markets

  • Criteria of an Indian War Chief

  • The power of low expectations. My early startuping. Ukraine.

  • Took almost a week and a half to recover from overtraining.

  • More thoughts on the Pareto Principle. Russian defectors and a single business email.

  • The Mother Russia of All Coordination Problems

  • Coordinating against an authoritarian regime

  • Why do totalitarian states restrict emigration? Immigration makes sense.

  • Practical tips for breaking junk media addiction

  • Morovec's Paradox means AI before robotics. Below the API.

  • News and Politics on Less Wrong

  • Sieges and Scorched Earth. Actionable information.

  • Operation Cyclone

  • Geopolitics on LW

  • Stories out of context: Rats in the bathroom. Library guillotine.

  • Wow. Spider-Man No Way Home was so good.

  • Improv Comedy as Treatment for Autism

  • Why does Spider Man and the multiverse work? Because it's about possible futures a kid could grow up into.

  • Extracting Trust from Untrustworthy Data

  • Nuclear strategy

  • Infrared & camouflage

  • The Javelin Missile Launcher. Fire-and-forget.

  • In 200 years, short URLs will be priceless family heirlooms

  • Hungry in Taiwan

  • The True Meaning of Chinese New Year

  • Above and below the market-clearing price, also competence

  • In a zero-sum game, anything which is bad for your opponent is good for you

  • Lanchester's Laws

  • How thermal optics work

  • Fast response vs prediction.

  • Kaz

  • None of my sources for Ukraine are news outlets or reporters.

  • Baseline ignorable risk/uncertainty, reply to this

  • A monopoly on truth

  • Quoting out of context

  • Induction and Deduction. Death by a single cut applies to induction.

  • Calling nations by their location, and what it implies.

  • Advance preparation and general understanding as prevention against propaganda

  • How Faraday Cages work

  • Combining uncertainties (and the Law of Large Numbers)

  • Freezing self for reset

  • I don't spend too long analyzing social reality to determine object reality because it's better to just go to object reality.

  • Repeating numbers and geometric series

  • Criticism Doesn't Generalize

  • Bicycling and happiness. I dislike cars.

  • "Vegas Syndrome" Paris Syndrome + Vegas

  • Shanghai and Suburbia

  • Morovec's Paradox

  • Quirkiness vs Standardization

  • Bicycling and locality

  • There is always a chance threats will be carried out. Game theory.

  • The faster I write something the better it is.

  • List of personal fire alarms

  • Attacking the Enemy's Command and Control

  • AI before robotics

  • Honors Advanced Calculus

  • "Good Conformists." People who are good peoplle because they're conformists creep me out.

  • Contemporary Nuclear Strategy

  • Civilization has always sucked. Bold action to solve physical problems has always been inefficient and wasteful.

  • Hostile informatic environments

  • Bitcoin and free energy in Russia

  • Language is mostly "I'm doing "

  • Rapid response looks like predicting the future if the world is slow

  • Re: Case study in misleading media.

  • Personal blog: Not hating the Russian people.

  • Registering Binkov's prediction

  • Professionalization of the military. Less need of conscripts.

  • Appreciating the difficulty of medieval warfare by bicycling

  • "The Innovator's Dilemma and AGI" Cheetahs, birds and the Innovator's Dilemma

  • Re: "Thus I frame the heart of the conflict between different simulacra levels. The fight is between those who live primarily in simulacra levels one, where they interact with and attempt to model and alter physical reality, and those at simulacra levels three, where one cares mostly about signaling group memberships and loyalties, and often also level four where they end up using vibes and associations of various level three moves in an instinctual way that they have learned is associated with success, and lose all ability to reason or make plans in a coherent way at all, let alone model the physical world." My model is people who never thought at that Level 1 in the first place.

  • Counting sides in policy debates

  • Writing insane characters

  • Cooking Classes and Grocery Stores

  • Actions and Mere Words

  • Cultivating Impatience for things that matter

  • Excitement and Risk

  • "Buddhism" and Buddhism

  • Missionary urge comes from lack of self-confidence

  • Binkov is about to be tested in Ukraine

  • Re: Russian X-Risks

  • Opionions evaluated lazily. Should schools be forced to teach Arabic numerials? Should we bomb Agrabah?

  • Historians and Biographers

  • Stuff I historically do wrong: pushing myself when I should rest

  • "If I wasn't hungry then I would be happy."

  • Re: "Enlightenment is knowing that no one does anything" Shame and hatred are two sides of the same coin. They're both about guilt.

  • Correcting "life is suffering"

  • Re: Success is Not Success

  • Enlightenment is a Qualia-Level Phenomenon. Enlightenment is hard to explain because it operates at the level of qualia. Why is enlightenment so hard to put into words? Because it's qualia. Shatting shame and guilt.

  • On not knowing what you want.

  • The Public Security Museum

  • Alternatives to YouTube's engagement metrics.

  • Strategic analysis of Facebook

  • The incentive for minimum quality bars when producing media

  • How to tell what you really want. Something with no social support. Vegas. Hedge fund.

  • Zvi on Jane Street

  • The news has always been full of misinformation

  • Frank communication about uncertainty and trust builds credibility and trust.

  • One of the hard parts of meditation is jumpstarting belief. Inefficient market is same way.

  • I think I'm just suffering from post-enlightenment existential confusion

  • Good stories come from wanting stupid things.

  • The "unable to love stories" phase. Now I'm back. Appreciating the artistry of magic tricks.

  • "Twelve hundred miles for a kiss." "Worth it." from "Real Steel".

  • Looking for enlightenment is like trying to believe that 2+2=4. It breaks the Litany of Tarski.

  • Enlightenment is like eating chocolate vs tasting it. Someone telling you heroin feels good will not make you addicted to heroin.

  • "Should"

  • Science and Enlightenment

  • Reproducibility and the Dark Night. Also interesting that I went through the phases without any Vipassana.

  • Meeting people again after COVID and stream entry

  • Why enlightened individuals don't like talking about enlightenment

  • There's something there and it's the most boring thing in the world.

  • The value of blocks of time

  • Bullshit signal: not getting to the point.

  • Reviewing Game of Trades' performance $8,000 gold

  • Capping your losses with historical data to cap your confirmed risk

  • Fractal generalization

  • Writing software that can find errors in its training data

  • Politics on Less Wrong

  • Re: Intro to Brain-Like-AGI Safety

  • People create the conditions they need, even if those conditions are stupid. I like desperate struggles for survival.

  • Building space stations in KSP

  • Why D&D scares me but Improv Comedy doesn't

  • More thoughts on enlightenment. Everything toned down. Similar decisions but less stress. Weird to think into echos of patterns from before the incident.

  • Building Communities

  • Using Improv to treat Autism

  • Communication skills scale well, have high leverage, always useful, unlike manual skills.

  • Learning to program is all about breaking your machines

  • Writing Press Releases

  • Above and below the market demand. Tons of requests vs zero. Does it scale? If so you can handle all those requests. If not you just pick one.

  • Either do stuff that's intrinsically personal (like meditation and cycling) or super scalable (like writing and tech entrepreneurship). Avoid things which require you to pay time to play (like networking events). Dancing and improv are allowed because they're intrinsically valuable.

  • Trouble using the word "want". Maybe "choose" instead.

  • Maou-sama and Aqua-sama. Non-consequentialist moral frameworks.

  • "I write songs about ridin' tractors…from the comfort of a private jet."

  • Skill (execution) and application (targets)

  • "It was random" as explanation of last resort

  • Salsa dancing as unfakable test of someone's inclusive fitness

  • Can I go without advertising?

  • Unstable space stations in KSP and meditation's effect on the human brain.

  • Re: Prediction Markets Fail to Mooch

  • The flip side [0,1] of

  • Incurably Stupid. Comments that aren't worth explaining to.

  • Leverage causes Pareto distributions. You don't see some people living 1000× longer than others.

  • Schools are incentivized to make you want more school.

  • Teachers can't motivate. Can ruin motivation. Personal philosophy.

  • Everything is amoral. Someone says something like "coordination is good". I respond by using coordination for evil.

  • I just realized programming binges and videogame binges are the same thing. Programming is bad for me.

  • Making offers easier

  • Designing my own news feed

  • Employers are bad at measuring engineering talent

  • Lopsided pairing markets: dating websites, job recruiters on LinkedIn, conversation exchange. How to operate, on both sides. Famous people and email.

  • Strategic Bioweapons (race-based genocide)

  • Using prediction markets to fund proxy wars

  • Avoid unnecessary politics. For my "Your Prediction Markets Can Be Used Against You" post originally USA/China.

  • Prediction markets for government policy as a source or revenue

  • Manipulating Prediction Market Outcomes for Fun and Profit

  • Influenceable Prediction Markets and Spread

  • Prediction markets incentivize influencing cheap, unlikely outcomes. "Prediction Markets Incentivize Terrorism."

  • Using prediction markets to influence outcomes

  • Err on the side of audacity

  • Vagueness, precision over bets

  • Weird rights: self-liberation in Germany/Austria/Belgium/Mexico (right to break out of prison)

  • Enlightenment doesn't eliminate the "I should be working" urge.

  • Facebook for old people. Better future.

  • Large organizations never acknowledge fault

  • My cooking is really complicated. Working first, then iteration.

  • Food, tools with personality. Small town fantasy.

  • Evolving my recipes

  • Savagery and Civilization

  • The 20s are back

  • "For many decades, the precise taxonomic classification of the giant panda was under debate because it shares characteristics with both bears and raccoons. However, molecular studies indicate the giant panda is a true bear, part of the family Ursidae."

  • I write characters that are just me and people think they're nuts/crazy.

  • Medical care and death

  • If not news then what instead?

  • COVID-19 as the liberal 9/11

  • Re: On Bounded Distrust

  • What makes the "Questionable Content" webcomic work so well?

  • Protection agreements and democracy

  • Is there a limit to how much time I can utilize effectively? Of course there is. What is the limit and what do I do with the rest?

  • Living with unbounded utility functions

  • Corollary to Pareto: restricting media intake to not-quite-zero has almost no effect on enjoyment. News can go to zero because leakage.

  • Tim Ferris: What would this be like if it were easy?

  • Making myself take more time to write makes my writing worse. Forcing myself to program when I don't feel like it makes my software worse.

  • Better to create a category than to compete in one <-- Marketing tip from Tim Ferris

  • Leverage Causes Pareto Distributions. Fungibility and Transferability. Being a commodity eliminates your leverage. Do not let your time become commoditized. Learning low leverage because non-transferrable (Six Specializations) and because classrooms. Happiness: pain & pleasure long-tailed. Learning is reverse. 20% of work gets you 80% to competence.

  • Japanese Emperors restricting access to their persons

  • Replying to comments and email multiplies them

  • Beliefs are defined by bets

  • Writing software only under optimal conditions

  • Indentured servitude

  • My post-scarcity post makes it seem like I support free markets. Do I?

  • By "post-scarcity", I mean most of our resources are burned on non-utilitarian activities. We are still well below the physical limits in absolute terms.

  • Academics biased toward "don't do ". Businesspeople biased toward "do ".

  • Labor shortages are subjective. Think in absolutes. Write in absolutes.

  • Visualizing steep learning curves

  • Profit off of people who are wrong. Implicitly ignore their words.

  • My best writing, like my best code, is written fast.

  • Nobody Cares What I Believe

  • Taking the Pareto Principle seriously

  • Defining "efficient market"

  • Liking money and being vain are correlated because they're both about comparative status relative to other people.

  • Review "The Devil is a Part-Timer!". Similar to "Sword Art Online". Looks like shonen survival-adventure-competition but is actually a romance. Review Konosuba too.

  • Corona Discharge. Million volt? I love the smell of Ozone in the morning.

  • Synchronous or asynchronous. No in-between.

  • I don't get LinkedIn. It's like Facebook but worse.

  • The only stuff worth writing is stuff that looks wrong. Part of not saying "you're wrong".

  • What excites me?

  • Latinx Asian Teriyaki woman.

  • A Magician's Perspective on AI Alignment

  • AI alignment, tricking people and being a magician.

  • Paying for ideas is often bad idea. Working prototypes is better.

  • Paying money to impress others wastes money. Same is true of time.

  • Shock Experiment

  • No comments on my articles

  • PG on email. Tim Ferris has the solution to many of PG's problems.

  • Fake work (email), fake learning (news, web surfing), fake friends (YouTube), Tim Ferris. Salaried employees paid by the hour.

  • Umbrella story. Taiwan. Hungry.

  • Being straight. Programming a computer to be straight.

  • Happiness is turning off your phone. Metaphor for meditation.

  • I've got a thing for women who wear suits. Japan story just an excuse for that. , Tsunemori Akane, Helltaker

  • Re: Training Robust Corrigibility

  • I like the ARC's approach to alignment

  • Classics project: public-facing Art of War or similar for ACOUP.

  • Stuff you actually use is cheap

  • Communication batching: daily, weekly

  • I don't find people interesting. That extends to me. I don't find myself interesting either. That's why I don't write about myself.

  • Experiment: One week of writing. Nah, that's stupid. Maybe not.

  • Being autistic. Constructing model of human people. Avoid emphasizing the labels. Solve the problem.

  • Priceless Dog Poop

  • Write honestly because if you can't do that then you won't know who you are. Teenagers feeling strong but being unable to communicate.

  • Don't save anything for later (writing advice)

  • The Right to Bear Guillotines

  • Writing from the heart

  • We write numbers backwards

  • Being selective about the people I'm around

  • The purpose of writing is to inhibit clarity

  • Decide what you want before seeing the options

  • Liking bad movies and other bad things

  • Minimal list of items again

  • The liquidity spectrum

  • What skill feels like magic? Programming. Improv.

  • Writing from the heart without rambling

  • Limiting factor is execution, not ideas

  • There are no such thing as "business skills"

  • Business skills

  • Executive Lessons from Margin Call (2011)

  • Jeremy Irons in Margin Call (2011)

  • Hurry up and wait.

  • Zygomorph's Question why not just work at Google and donate 90% of income? Where does it end?

  • Decide what you want before looking at the options.

  • Peopleware

  • Do Not Resuscitate Me

  • Procrastination is a way for dealing with unreliable reality & people

  • Learning to negotiate. Dates. Business contracts. Has to be real. Gambling, communication, improvisation, walking away. Actually having options. Haggling sucks. You don't have to much in developed nations.

  • Make a chart of all dance moves

  • Stuff I wish I did at various ages.

  • How different work affects my personality, emotions. Nonfiction, fiction, software.

  • Officers eat last

  • Why orbital maneuvers require high s.

  • "One difficulty here, though, is that people are often mistaken about where they fall on the spectrum from conventional- to independent-minded. Conventional-minded people don't like to think of themselves as conventional-minded. And in any case, it genuinely feels to them as if they make up their own minds about everything. It's just a coincidence that their beliefs are identical to their peers'." Looks like everyone is right. "You know that some accepted ideas will later turn out to be wrong. See if you can guess which."

  • Leadership lessons from the Indian UN lady in The Expanse

  • Transparent terminals

  • Future is more interesting than simple universal explanations.

  • Transparent holograms projections will always be the science fiction

  • My favorite thing in the world is building systems and then forgetting about them.

  • Why learn to write? Because most of the communication will be writing and speaking. When you give orders, much of it will be in writing, especially if you're trying to give them efficiently. Asynchronous. As a rule-of-thumb, synchronous communication is done verbally, asynchronous communication is done via writing.

  • "There will always be..." MIB quote

  • Optimizing for excitement. Hogwarts is full of evil wizards and monsters and traps. The same goes for generic high fantasy. Without risk there is no excitement. The trick is to have the right challenge level. Somebody is wrong on the Internet.

  • Having things worth writing about is harder than the technical aspect of writing

  • The job-hunting chicken and egg.

  • Bringing back terminals. tmux and mosh

  • Take all your theories of how to make an AGI and dump them into a nonfiction post

  • I don't spend enough money. How to do the opposite?

  • How do I find good quotes? I've read PG so many times I just remember them.

  • Preserving a sense of wonder

  • Teaching a middle school class

  • Cure for people saying negative things about you. Read negative reviews of stuff you like.

  • Hard to refute bullshit. Even harder to refute negativity promoting inaction.

  • Why are so many trans women elite programmers?

  • Contacting people when you're young

  • The Advantages of Word-of-Mouth

  • Ishuzoku Reviewers Chapter 38. Diversity, inclusivity and being true to yourself.

  • Satire about patriotism making fun about how readers aren't all American.

  • At Home in Japan

  • Breaking the First Commandment

  • When I'm feeling bad about comments I read the comments for The Matrix Resurrections and Johnny English.

  • The virtue of being totally upfront with people

  • Heuristics for inefficiency. New systems. Skill required.

  • Looking silly, especially when lots of people are watching

  • Choosing the terms of engagement

  • Premature Legibility. Premature optimization vs altruism.

  • Alternate prehistory fanfiction

  • Infinite suffering. Infinite good. How Not to Save a Million Lives is the hardest post I've ever written. Because I have a heart. Do not blame others until you have stood in their shoes, holding the vanguard's shield against the forces of evil.

  • Carrying the team.

  • Ups and downs of startuping "What Surprised Me About Entrepreneurship" and "How to Not Save a Million Lives"

  • Long-tailed altruism

  • How to save a million lives. Bite-detection software.

  • "I don't mean to be racist but…" being asked in Chinese conversation exchange. Binary races.

  • "Deriving Social Intuition from First Principles". Story about autism.

  • Socializing from books

  • "Relics of a More Civilized Age". Learning Linux. "On the Shoulders of Giants".

  • Realism and idealism

  • Diversity and Rigor

  • Human language mostly built on top of status sense

  • "and they think this is okay" from Joel Spolsky. Medieval fantasy.

  • So happy to be learning new tools. tmux and mosh. I feel like I did when I first learned computers except now I'm actually good at it.

  • Cats are not Dogs

  • Is there a bottleneck? More or better.

  • Kinds of people to select as friends. Limited time at older age due to better utilization.

  • Haters, criticizing others and not getting anything done yourself. Politics does need people to criticize. Critizing others for stuff you've never done.

  • The 20% of people who give you 80% of problems.

  • Positivity in relationships

  • Against complaining

  • Jobs where experience does & doesn't matter.

  • Tell past self not to worry about not having skills to do stuff

  • Winning Arguments vs Winning Contests. Winning a few important minority bets. Deviation from consensus makes you look wrong.

  • Everything is a Race [because lifetime is a clock]

  • Why we write numbers backwards

  • Re: How to Disagree

  • Problem Space

  • Keep LW Weird. Oppose congealing. Wrong is bad, but exploration and experimentation are good. Dynamic. Draw the line at rudeness and anti-epistemic literal cults.

  • Things I don't write about: [redacted]

  • Sepoys using greased cartridges on British troops

  • I take notes in meetings but I never read them. Same as in classes.

  • You get rich by owning things -- Sam Altman

  • Strategic predictions about ML

  • Some sort of misalignment contest

  • Clients don't know they're bad communicators.

  • Magical volunteers

  • Skill Trees

  • Human values are trash

  • LW and experts online

  • SMBC about flu

  • Learning is a race

  • Facts vs Patterns

  • Maximizing Transferrence

  • Mundane mind control. Junk media.

  • I think sales has a bad reputation because of sampling bias

  • It's not that everything turns into sales. It's that everything turns into communication.

  • Communication for autistic people

  • Prospecting for invisible problems

  • i3 tutorial

  • Convergence between tmux and i3

  • I should practice writing idea handles. Yeah, that's what I should be doing. More designing memorable titles. I haven't done stuff like this since Antimemetics. One approach would be to be the guy who is hopeful about AI.

  • Street magic and manipulating the energy of a room

  • "Analogies are good for explaining concepts for the first time. But they have no value in debate. Analogies are not logic, and they are not relevant facts. An analogy is literally just two things that remind you of each other on at least one dimension. When I see a cauliflower, it reminds me of a human brain, but that doesn’t mean you should eat brains in your salad. When your debate opponents retreat to analogies, it is because they have no rational arguments. You won." from Scott Adams' Win Bigly

  • Cognitive dissonance = doublethink

  • Stupid arbitrary goals. Benefit of orthogonality.

  • Book review of How to Win Friends and Influence People

  • Intimidation, to balance out the sales post

  • Good sales looks like customer service. Most people see the worst. Opposite of media where everyone sees the best. Project designer. Easier to keep a customer than find a new one. Large purchases. Large purchase dynamics.

  • The Japanese financial crisis

  • New media platforms

  • Betting on black swans

  • Explain biases of science fiction. Start by analyzing serious hard sci-fi and then switch to explaining Date A Live.

  • My first insight cycle. MCTB. Precise falsifiable predictions. This would be a valuable post.

  • Business lessons from magicians. Near-perfect correlation between skill and money.

  • Making Marvel sequals that work. Put the character in another genre.

  • The ethical questions I struggle with

  • My personal, favorite approaches to building AGI

  • Tell true stories

  • Triaging comments

  • "I don't think I'm that famous." ―Tim Ferris

  • Looking Stupid Quota

  • Re: Dealing with haters "I would take this one step further…. Not only do you want to be content to be thought foolish and stupid…you want to practice being foolish and stupid.

  • Wuwei and the Copenhagen Interpretation of Ethics

  • "The Right Flavor of Crazy". Readers should question whether you're sane. But concision is important too.

  • Chuck Palahniuk on building war memorials before the war page 23

  • Try to Agree. Mostly social businessy thing. Also relevant to multiple models.

  • What to do instead of arguing.

  • Write in absolutes.

  • Tell stories. Response to my previous

  • The one day the newspaper was blank was when something important happened. It was after 9-11.

  • Communicate Concisely

  • Nonperishable knowledge, response to perishable knowledge Why books matter.

  • Practice on hard problems. I learned about entropy. I do free business consulting for penpals in Shenzhen. It's a good way to practice my Chinese and business skills.

  • Mandatory information in language. Singular plural Chinese people.

  • Clubhouse retrospective

  • Disaster & Opportunity Checklist

  • How few people showed up to Chuck Palahniuk's and Brandon Sanderson's early readings

  • Why polished boots

  • Designing a custom product for a client's ill-defined needs

  • "The Man Outside the Arena" goes on my personal blog.

  • Guide to Samsara

  • Statistical mechanics for information theorists. AI researcher who literally doesn't know the mathematical definition of entropy.

  • What executive function feels like

  • The Wolf from Pulp Fiction

  • Asking to get paid in a disorganized operation.

  • The git push story of data collection.

  • Stateless Safety

  • rm -rf, guns and motorcycles

  • Answering Chinese in Chinese. No word for "yes".

  • Writers' mind as separate. Learning to trust it.

  • Containing Consequentialism: Power-maximizing a stateless AI. Should be science fiction about Bayeswatch.

  • A few fundamentally different AI architectures. Consequentialist. Stateless. Complexity-minimizing.

  • Choosing to not save a million lives

  • Under the Hood of a Non-Consequentialist AI

  • Locality in software

  • Eitan's Jewish Space Laser and von Neumann machines

  • Kids believe in monsters ridiculous. Actually how it worked.

  • Cold lights, cyberpunk, and my crusade against the color blue.

  • What yellow used to be. Wood, grain and skin.

  • My working conditions. I like working in basements at night because natural light varies too much.

  • Problems of objective-based AI alignment similar to overmanaging human beings.

  • Gender-segregated food. Sushi and Italian pasta.

  • People like to work at their level of competence

  • Beating Parkinson's Law

  • Stupid comments and stupid downvotes.

  • "Write for the Audience you Want". Writing on Less Wrong is has dramatically increased the quality of conversation I'm used to.

  • Functional Programming

  • Why it tales longer to move large limbs. Mathematical analysis.

  • Gravitational pull inside the Earth.

  • Easy to play chess. Hard to recognize a chessboard. Haven't yet figured out how to walk to a chessboard.

  • Muscular hydrostats

  • Case for free information. Was going to quote The Most Powerful Idea in the World but Overdrive wouldn't let me search it.

  • Identifying exceptional people.

  • What's the smallest bomb that could destroy the Earth?

  • Cheat Codes or Well-Designed world

  • Ask their servants and wives

  • "Think of the history of data access strategies to come out of Microsoft. ODBC, RDO, DAO, ADO, OLEDB, now ADO.NET – All New! Are these technological imperatives? The result of an incompetent design group that needs to reinvent data access every goddamn year? (That’s probably it, actually.) But the end result is just cover fire. The competition has no choice but to spend all their time porting and keeping up, time that they can’t spend writing new features. Look closely at the software landscape. The companies that do well are the ones who rely least on big companies and don’t have to spend all their cycles catching up and reimplementing and fixing bugs that crop up only on Windows XP. The companies who stumble are the ones who spend too much time reading tea leaves to figure out the future direction of Microsoft. People get worried about .NET and decide to rewrite their whole architecture for .NET because they think they have to. Microsoft is shooting at you, and it’s just cover fire so that they can move forward and you can’t, because this is how the game is played, Bubby."

  • Use Microsoft vs Web for innovator's dilemma

  • If startups write software and software is eating the world, including military stuff then does that mean startups will eat the military?

  • Case studies in software business strategy

  • Software technologies I love and hate

  • Why I gave up on mobile app development.

  • "Unfortunately, for most projects, you have to decide on which world to use before you can write the first line of code, which is precisely the moment when you have the least information."

  • More management methods extending Joel Spolsky's list

  • Wealth prevents conflict living together.

  • Cats don't want to go to the vet. Have you seen what happens when we try to vaccinate human populations?

  • Basically indifferent to suffering

  • Don't argue because it's a way of verbally bullying someone into doing something.

  • Manager Schedule. You are a bottleneck. Sooner matters.

  • One thing I love about biblical stories is how they are totally indifferent to modern sensibilities. Antimemetics. Folktales from Iraq.

  • Light and Darkness in the East and West. Light Yagami. Seto Kaiba. Fifty Shades of Grey. Dark eyes. Ones and zeros in a computer. Attack indirectly the problem of white people mistaking whiteness for universality.

  • Yin and Yang. Water and Fire.

  • Daoist Value Functions

  • Protect yourself from criticism under the cover of censorship.

  • YouTube removing likes. Media control. "Star Wars: Galactic Starcruiser."

  • For a boss, doing something sooner is equivalent to an employee doing that timedelta of work.

  • "I believe in you" from boss.

  • Give orders concisely

  • Minimum time consumption of various synchronous and asynchronous communication norms

  • Concise instructions. Clear goals. Executive.

  • Sequence on bottlenecks. Bacterium talks to a person [can be illustrated].

  • Orbital mechanics

  • Same voice reader as Brandon Sanderson. Daniel Ingram. Personal blog.

  • Natural Amphetamines and Algernon's Law

  • Going out when it's really cold.

  • Make offers even if they're slightly wrong.

  • "I'm not ready." "No one ever is." from Doctor Strange

  • Acting natural while doing magic

  • The Japanese magic museum

  • Re: my imaginary hedge fund

  • Teaching myself to program

  • Cascading Technologies

  • Stories: Heart of Darkness, Pilgrimage to Las Vegas, Ten Minutes with Sam Altman

  • Advice for smart people

  • Robot Park. Mix it with a romance?

  • Recommendation Algorithms

  • Thoughts on YouTube changes

  • With the conviction of gravity.

  • Re: "Writing and Speaking". Call it "Speaking and Writing" with the "Re:" in the first line. MLKJ. Barack Obama. John Greene. Tim Ferris.

  • Conviction

  • Bribing educators

  • Age of picking heroes

  • Vegas. My Vegas adventure.

  • COVID response story

  • The Nigerian bankers

  • 🐻 + 😺 = 🐼

  • Coordination and Subjugation

  • State-of-the-art no longer impressive

  • Example of a hy macro: pbar

  • Review of Death Note and Fullmetal Alchemist

  • Lying about reasons for doing things

  • I think writing software is bad for me.

  • It's not a priority. (Taboo "I don't have enough time".)

  • Heuristics for Scalable Interactions. Bulky metaphor.

  • Seeking out for Discomfort

  • Young generation on overthrowing AI Safety authority

  • Post-Stealth Combat Aircraft

  • Bullishness on AI risk

  • When will WallMart or McDonalds managers become automated? Zvi's below the API.

  • Reverse concrete vs abstract. Free speech.

  • Satirical post about how we need to eliminate disagreement on LW. Tightly narrow what is acceptable.

  • Telling technology what we want instead of how to do it

  • and cyan

  • Generalizing to Too Low a Resolution

  • My teacher who left his sanity in West Africa

  • Doom: Annihilation review

  • Autonomous tanks

  • Cloud Atlas and Jordan Peterson

  • Joel Spolsky vs Paul Graham on back button page hack. Often favorite essays are debates frozen in time.

  • I like writing because it's like lecturing at people and nobody can stop me.

  • Middle school impressed by news. What don't understand is that my company put those stories into the newspaper. Didn't have a PR team. We did it ourselves. Credibility didn't break when reading stories. It broken when I helped write them. Never believe your own propaganda.

  • Favorite posts of readers are when I write about personal experience.

  • [NSFW] Review Interspecies Reviewers

  • Review of Tahir Shah's works. Unplacable genres like Willie's Wonderland and [NSFW] Ishuzoku Reviewers. I love things that don't fit into established categories. Motorcycle story.

  • Dating networks on my homeworld psychometrics. Dating pairers. Paid for results. Goodhart's law on IQ tests.

  • Quick overview of Confucianism, Daoism and Legalism

  • Against consistent worldviews

  • Animal Brains as Homeostatic Stabilizers. Western goal-oriented utility maximizers. Eastern process flow operators. Internal harmony. External maximizers.

  • Non-hostile superintelligence. Gears level AGI.

  • I don't pander because pandering inhibits my truthful thinking.

  • The Quirky AI Hypothesis

  • Escape hatch corrigibility Matrix story

  • Foraging for coconuts story

  • Airsoft on my homeworld. Comfortable concrete bunkers. Equipment is handicap. Rubber knives. Usual steal an enemy weapon. 1 vs many. Combat realism. Boredom punctuated by extreme fear. Friend of mine on Earth who didn't like Girl Scouts. Walking at night without flashlights. Never used UCP. In the big tournaments there were supply drops to mix things up. Basically the Hunger Games except nonlethal and not evil. Rangefinders adjust muzzle velocity.

  • Eutopia is Dangerous

  • Europeans invented ships. Tremendous wealth, lots of ideas and a severe labor shortage.

  • How I would build an AGI

  • Endeavor to copy someone's art and then get bored before reaching that objective.

  • Gifting precious artifacts.

  • Put me somewhere else clip

  • Primitive Gift Economies

  • Mad Max Fury Road and design. Red Notice.

  • What does(n't) matter in high school and college

  • Cyberblitzputsch

  • Pro-China satire about AI alignment coordination

  • Turn the Other Cheek is an extension of Help Your Enemies When They Are Numerous

  • How a superintelligence can easily take over the world. No need for nanotech and von Neumann machines.

  • Non-world-optimizing Superintelligence. Simplicity instead of world optimization. Error-entropy.

  • Civilizational Bottlenecks

  • Intelligentization and preventing insubordination

  • Truth as a core value

  • "There is always a tendency for rich customers to buy expensive solutions, even when cheap solutions are better, because the people offering expensive solutions can spend more to sell them." also 50K software joel on software

  • Prediction-based court system

  • Civilization and Penis Length

  • Roadmap to AGI

  • Protecting trade routes with China...from China

  • When the cards are down, second time.

  • Literal and metaphorical truth JP: Hell is bad is not a factual claim (paraphrased) Not how minds work either. Don't agree with either of them. Sam Harris on suffering. Jordan Peterson on metaphorical truth. Neither traditional materialist. 1:27:23 Jordan Peterson's God. Atheist Judaism, Zen Buddhism, Protestant Christianity. You know who the only people in my personal life are who take religious texts literally? It's the atheists. Also Sunzi because it's literally a war manual.

  • Structure of rationalism. Free speech first. Then science & math. Scientific establishment. Then material reductionism. Not a historical claim.

  • When I was a kid I really liked turquoise but I couldn't explain why.

  • Lawful Evil Paladin from D&D gets interrogated for acting ethically and must defend herself.

  • Robust Control Systems for Autononomous Weapons

  • I frequently assume that just because I can do something means other people can too. This is a reasonable prior for most people but an absurd one if you are weird.

  • Dynamite Fishing for Readers. That which can be destroyed by the truth should be.

  • Re: Jhanas and the Dark Room Problem

  • Be skeptical of arguments for inaction

  • Footnotes are symptomatic of bad writing

  • Classic scifi tour

  • Emma Stone

  • AI Alignment for Military AIs

  • Human Ethics for Killer Robots

  • Speed of easy stuff

  • Re: Finding Great Developers

  • The ancestors scene from Mulan

  • True to Yourself

  • Saffron from Firefly and Wichita from Zombieland

  • Speed is victory

  • Can we cure aging?

  • Music and storytelling adding to a pattern

  • Against the Current (band)

  • Copying vs Understanding

  • Introduction to Programming in Hy

  • Special Relativity from first principles

  • adversarial networks and suffering

  • Light and Darkness in Yu-gi-oh and Death Note

  • Opt-In Eugenics

  • China's space program

  • If your ideas are any good you'll have to shove them down others' throats

  • What if there was full nuclear exchange?

  • I think the idea someone might accidentally build a misaligned superintelligence is ridiculous. If a misaligned AI takes over the world it'll be deliberate.

  • So-called Infohazards are a cult tactic

  • Science without experiment not science. AI safety.

  • John Scalzi: dinosaurs, mammals and cockroaches

  • Make precise predictions

  • History of Irish conflict and the Troubles

  • How do you teach independent thinking?

  • Better at acting normal when I'm sick. Dumb, unmotivated and my senses are dulled. Makes it easier to model other people.

  • Extraversion, Dopamine and Cocaine

  • Snow camping

  • Laser power transfer, near infrared

  • Counter propaganda. Different from series on well-meaning but wrong.

  • Interview with A'gong

  • On being a "real" or "professional" magician/author/whatever.

  • Readership survey with bad multiple choice questions

  • Writer's block Title: "Publish Crap"

  • Regression to the mean

  • Stupid comments you get while writing

  • Organization can't operate if everyone has ethical qualms.

  • Jumping from one layer of abstraction to another.

  • Destructive warfare effect on ethics

  • Character design practice

  • Various character analyses

  • Sword Art Online

  • Alternate history without uranium

  • How society should adjust for genetic knowledge anathema until mandatory: embryo selection & genetic engineering of children. Infanticide should be legal.

  • Wachowskis never feel like they're phoning it in

  • Studying anthropology to understand how ethics are pretty consistently determined by environmental constraints

  • Pretty used to realizing that we're caused by evolution. What about when your ethics are caused by social Darwinism?

  • Catherine the Great's attempt at liberalizing Russia

  • Tax evasion via art. How does it work? Something to do with manipulating process for tax writeoffs.

  • Art auctions & public knowledge. Private not. Buyers and sellers both encouraged to raise prices.

  • AI before robotics

  • Programmed theory of ageing

  • Octopus Suicide

  • Moonshots

  • On getting feedback

  • Conditions for systemic abuse: Extreme power differentials. Secrecy. Closed to outside ideas. Lack of fluidity. Anti-epistemic proclivities.

  • How to pretend to speak Chinese? How good is your Chinese?

  • YouTube uses a generative algorithm to make new videos

  • Comedy rides the line

  • I'm confused about mesa optimizers

  • Entropy and Bayes Equation

  • Don't Refute Bullshit

  • Secrets are the Enemy of Truth

  • Infohazard Policies, Responsible Disclosure

  • Underspecified Abstractions. "I don't know" is for questions. Abstractions vs reality.

  • The Autistic's Guide to Normal Human Behavior

  • Enough people talking about Westernism and American hegemony and liberalism as if it's world peace―enough that it's become a pattern. And yeah, peace is a result. But it got there with violence and force.

  • Make your own hard times

  • Taiwan last soldier of WWII Teruo Nakamura

  • Cultural appropriation

  • Greased lightning and the breast size comparison scene from Railgun

  • Fix it Yourself (or at least send an issue report)

  • Rationalist vampire story

  • making fun of myself

  • 呢/ね/eh

  • I'm just a _

  • そう

  • Where's my magic sword?

  • The value of fiction

  • You can't save everybody.

  • Just do stuff

  • Modern Art, response to whither tartaria

  • Preceptual Reductionism

  • Review several drawing books including "How to Draw: Drawing and Sketching Objects and Environments From your Imagination"

  • History as arrows in debates

  • Systematic extermination of competing cultures, Coordination via Conquest

  • "Protected Health Care Room" vs "Quarantine Zone"

  • Beware the Labyrinth of Abstraction. Words should connect to reality. Akrasia is inevitable byproduct of living in the Imaginary Land of Abstraction. Not how we learn language. "". よろしくお願い致します. https://learnjapanesepod.com/learn-japanese-pod-02-how-to-do-a-self-introduction-in-japanese/ 04:54 "Basically, the easy way is just say it at the end of your self-introduction and don't woryr about it. よろしくお願い致します。That's my rubbish quick explanation…. But don't worry about it too much. Just use it at the end of an introduction or at the beginning of an interaction when you're going to be working together with someone. So how is that for a terrible explanation? But just remember that 'よろしくお願い致します'―That's the polite version. Or, more commonly, when you know the people better, you'd say 'よろしくお願いします'." Ends at 06:43. Chanting at meditation. At this point it's like the "ta ta ta" of an acapella. https://youtu.be/JseWhrUz9TY The Far Side dog barking https://i.pinimg.com/originals/1d/b5/27/1db5279f8a07f1601bf75a92caf9db0a.jpg Taking over the world conversation with Spencer https://www.reddit.com/r/WinStupidPrizes/comments/pmcxk0/what_did_he_think_what_gonna_happen/

  • How I think about Social Norms

  • Grand Social Theories with No Historical Context

  • Culture as Coordination

  • Simulation in a simulation

  • Modern art and how technology unbundles things

  • Generalizing from scifi generally overprices how much things cost.

  • Automating Tim Ferris

  • RPGs, sanity waterline, practical.

  • Punishment and Reward are Asymmetrical

  • Rick and Morty and Jordan Peterson

  • Is a field just the history of the field. If so then no progress.

  • Book Review Moonwalking with Einstein

  • Satiricial argument against people saying something ridiculous except the ridiculous thing is from an alternate world.

  • Thoughts on CSHW

  • Re-do the antimemetics sequence. Maybe re-title it "Extrapolation".

  • The Monkey in My Closet. When to determine when there really is a monkey in your closet. How to know when you're right when odds fall below the lizardman constant.

  • The Nihilism Fallacy

  • GPT-3 tarot reading

  • All writing advice can be ignored

  • Write statelessly

  • Guide to snobbery and pretension

  • Fiction is about conflict

  • Learning is Always Active

  • The antidote to pessimism is victory

  • Developmental milestones SSC. Differentiating "some people" from "everyone".

  • Premises for realistic hard scifi. Fire control systems. AI more than robotics. No FTL. Use wheels. No hovercars. No artifical gravity.

  • Can China match US innovation

  • Bioterrorism is Coming. If I was a supervillain…

  • Whenever I can communicate to a kid without saying things they prefer it.

  • [Learn to ask for what you want.](https://blog.samaltman.com/startup-advice-briefly Hard because efficient market.)

  • First time alone in Taiwan

  • Re: Advice for Ambitious 19-Year-Olds

  • Strong AI default to unprecedented concentrations of capital and power

  • Famous person [redacted] said in an off-the-record meeting he would like to be asked: How should I figure out what to do with my life?

  • What could an AGI invent that would change the world? Broadest impact (or benefit) for humanity.

  • Trends going so we can only train big models a few time.

  • Deconstructing compassion via material reduction

  • How do you make an AI compassionate?

  • Are intelligence and consciousness separable? What do I [Lsusr] think "consciousness" is? How would I define it? How do you define "intelligence"?

  • Is it worth pursuing non-AI things when AGI could solve everything?

  • Are nuclear energy companies near breakthroughs?

  • Adversarial training data. Based around small data. Real project. Slow takeoff. Alignment and implementation same thing in this context. What is a bite? "The Gesture Detection Problem".

  • AI from kids. Learning how they approach the world.

  • Pixels and parameters

  • Human-level doesn't matter. Computers won at chess decades ago. If everything then it doesn't matter. They follow a different progression.

  • Several people advocating laws passed on AI. Scary without defined.

  • Different kinds of responsibility

  • Satirical post about how I try to explain why I treat humans and dogs differently. Discrimination.

  • Skipping the intro

  • Edge of Tomorrow

  • Noragami

  • Cavalry charges

  • The Economics of [redacted]

  • Suppose you wanted to measure differences in intelligence across races. How would you even do that?

  • Manufacturing serendipity

  • How does stealth technology work?

  • Non-agentic AI

  • Comedians as canaries in the coal mine

  • What's causing all this World Peace?

  • An Era of Unprecedented Free Speech

  • How to actually give up junk media

  • Disruptive inventions I wouldn't have seen coming

  • How hard it used to be to get access to information

  • Every way to die

  • Lessons from magic tricks

  • Changing my mind about exercise

  • The monkey in my closet

  • Handling Pokemon scope creep

  • Change in values over time. Stuff I don't care about like shiny pokeballs. Not just toys.

  • The current military situation of ISIS and the Taliban

  • AI is the mind killer

  • Everything I wished I knew job-hunting

  • I don't know why I like drawing

  • Moonshots are for charity ["Barbara Fried puts it this way. “If you’re earning money for personal consumption, there is a very steep, declining marginal utility of income. After your fifth Porsche, do you really need a sixth? . . . But if you’re earning money to give it away to charity, there’s no diminishing marginal utility to money. The last life you save is worth as much as the first life you save.”"](https://finance.yahoo.com/news/ftx-ceo-sam-bankman-fried-profile-085444366.html How I think about money)

  • Various graphs of quality of live over last century

  • The doing nothing failure mode of trying to do good work

  • Helping out with refugees

  • Was feeling awful. Sat down and wrote code. Bicycling. Reading books on weightlifting.

  • Happiness is having more money than other people. My reference class is historical and international.

  • Re: People call me optimistic. Also pessimistic.

  • Against the scaling hypothesis

  • Japanese perspective of Pearl Harbor

  • Cultivating different kinds of comments

  • Research into stealth technology

  • Using the word "bullshit"

  • Multiple maximums. Private jet vs first class.

  • In every (most?) domains you have a few people who know what's going on and everyone else has just memorized a script

  • "Can we talk about something else, please?? About ready to hit the delete button on this app again 😂💉😷" My guess was the fall of Kabul to the Taliban, but that doesn't make in context.

  • I systematically overestimate what other people can do.

  • Moral systems very different from our own

  • "Scott doesn’t explicitly ask the obvious question, which is “does school actually do anything useful while it’s traumatizing kids to obey authority rather than examine physical reality, and ending their lives one minute at a time?” or alternatively “But Can Kids Recover From Not Missing Even Quite a Lot of School?” but he does point out that school didn’t teach him anything useful and now he’s Scott Alexander, and it didn’t teach me much of anything either, so search your experiences and draw your own conclusions, and then draw your own secondary conclusions about the pandemic."

  • Beyond the reach of science. Cannot generalize from historical data. Hostile learning environments. Nonconvergent series.

  • Context-dependent integrity

  • Environment as object Truth is universal. Outsider perspective.

  • Beyond the reach of science. Lack of research on LSD.

  • Unpredictable points in technological progress

  • Effect of transformer models on the media ecosystem

  • Re: [Daniel's median future](https://www.lesswrong.com/posts/6Xgy6CAf2jqHhynHL/what-2026-looks-like-daniel-s-median-future#What_about_all_that_AI_powered_propaganda_mentioned_earlier__ https://www.lesswrong.com/posts/jusSrXEAsiqehBsmh/vignettes-workshop-ai-impacts)

  • Leadership = taking responsibility for something greater than yourself

  • Impact of ML on society

  • Asian representation

  • Money as a lost cause

  • Coming up with ambitions

  • Prediction: self-driving cars driving around in circles instead of parking.

  • Creating a luxury experience. I can't think of a book on the subject.

  • Creating luxury experiences "Nonstop Dan" videos

  • How what matters changes in different classes. Capital for the rich. Intelligence/credentials for middle class. Tools for poor.

  • Normal subgroups

  • There is nothing noble in being superior to your fellow man; true nobility is being superior to your former self. ― Ernest Hemingway

  • Smart kid topics: math to study, requirements to be tech founder, integrity and doing your chore, adults don't get to flake—founders less so, premature optimization and focusing too hard on tech

  • Wax on, wax off

  • How I measure intelligence. A Word to the Wise

  • The low fidelity mistake

  • Love interests don't struggle. Ex. HSM.

  • The value of first impressions

  • Telling superior people they're normal is just as bad as telling handicapped people they're normal.

  • The future of AI

  • Accidental risks vs bad actors

  • Meeting people on the Internet

  • Scale

  • Leverage

  • Forever setting your sights higher

  • Does this person energize me?

  • Capitalist mindset

  • The "stealing" equivalence. If "I hope they steal mine" is strictly better then you shouldn't use the word stealing.

45

6 comments, sorted by Click to highlight new comments since: Today at 4:26 PM
New Comment

Farming long-tailed upside, Manufacturing Serendipity

I've been thinking about these ideas for a while after reading Taleb, and I'd like to signal that demand exists for posts like this. (I assume you've read PG's Black Swan Farming)

My personal, favorite approaches to building AGI

This strikes me as a terrible idea for a post. Even if you're bearish on x-risk, why give up your edge?

Why are so many trans women elite programmers?

Being a trans woman is a correlate of autism, and estrogen has been shown to reverse autistic symptoms. Maybe going through early brain development while autistic, and then later toning down the autism via estrogen, somehow results in the kind of brain that's really good at writing software.

I'm biased towards all the prediction market ones, naturally haha. In case you wanted to get a head start on manipulating markets for fun & profit:

Thanks for posting this, I like reading lists of people's ideas.

Posts I'm most interested in reading:

  • Aluminum fuel cells
  • Re: nation-states House-elves
  • Nuclear strategy
  • Contemporary Nuclear Strategy
  • Corona Discharge. Million volt? I love the smell of Ozone in the morning.
  • A few fundamentally different AI architectures. Consequentialist. Stateless. Complexity-minimizing.
  • Natural Amphetamines and Algernon’s Law
  • Non-hostile superintelligence. Gears level AGI.
  • How a superintelligence can easily take over the world. No need for nanotech and von Neumann machines.
  • Laser power transfer, near infrared
  • Premises for realistic hard scifi. Fire control systems. AI more than robotics. No FTL. Use wheels. No hovercars. No artifical gravity.
  • What could an AGI invent that would change the world? Broadest impact (or benefit) for humanity.
  • Anything on prediction markets

This list seems to contain some duplicates:

Also

  • Samsara
  • Guide to Samsara

Yeah. It's a lightly-edited version of the text file where I jot down all my ideas for posts.

Advice for Smart People with Autism

Even if you never write this one (or any of the other ideas with "autis" in the title), I'd love a 1-3 sentence summary of your suggestions.