My latest paper about ChatGPT: Does ChatGPT Understand the Concept of Tragedy? Symbolic AI & Neural Nets
Abstract: Tragedy is an abstract concept. David Hays proposed that abstract concepts can be metalingually defined. A term, such as tragedy, can be given its meaning by a pattern over a string of words. By using a cognitive network to define tragedy Brian Phillips investigated that computationally. In this paper I have taken Brian Phillips’ definition and presented it to ChatGPT in six different trials, where each trial also included a short story. Three stories satisfied the definition and three did not. ChatGPT was asked to indicate whether or not the story satisfied the definition it was given. It was correct in each case.
Anyone else find it weird how OpenAI goes to frankly absurd lengths to be "unbiased" (including flat out lying about certain sensitive topics), presumably to avoid pissing off the wrong people and getting negative attention for OpenAI and its services. But then, the CEO of OpenAI has as his twitter bio a symbol of his ethnoreligious identity, which he set this as in the midst of one of the bloodiest and brutal ethnoreligious conflicts in modern history, one which is amongst the most sensitive political topics in the world right now. Really gets the noggin joggin.
Has anyone else noticed Google Doc's autocorrect get much more aggressive and stupider recently?
Previously, it would just attempt to correct mispellings or grammar errors. Sometimes it would get one wrong, but for the most part it was ok, and it would never attempt to apply the same correction more than once if you backspaced and changed it back. However, perhaps a few weeks ago, it got a lot more aggressive and also stupider, and now frequently makes changes that aren't even gramatically correct.
For example, when I was writing just now, in the space of one paragraph, it changed the "worn" to "worms" in "However, there were also small holes worn", and it changed "I applied" to "I apple".
Furthermore, it no longer gives up if you undo its correction. I've seen it apply the exact same (mis)correction three or more times when I backspace and fix the "correction".
Does anyone here know what's going on here? I'm curious what people on Google Docs would say about this.
He's an update on the kid-prompted, generative-AI-generated, smart-phone distributed kiddie porn saga at Beverly Hills middle school. Five of the 16 perps (err students )involved have been expelled under a stipulated agreement with the parents of the perps (so the parents have accepted the expulsion and they won't legally challenge the school district). It's unclear what if anything is going to happen to the other 11 perps (err students) involved.
The county DA and BHPD are still investigating, so criminal charges could still be filed. But... "California’s laws against possessing child pornography and sharing nonconsensual nude pictures do not specifically apply to AI-generated images, which legal experts say would pose a problem for prosecutors."
It's unclear if the parents of the victims of this AI-generated classmate porn will sue the parents of the perps or the school district.
Walk & Talk: We usually have an hour-long walk and talk after the meeting starts. Two mini-malls with hot takeout food are readily accessible nearby. Search for Gelson's or Pavilions in the zip code 92660.
Share a Surprise: Tell the group about something unexpected that changed your perspective on the universe.
Future Direction Ideas: Contribute ideas for the group's future direction, including topics, meeting types, activities, etc.
I am a somewhat intelligent person -- I once measured with a 143 IQ, which seems a fair number to capture how I qualitatively think of my intelligence. Is there any fast path to a decently paying job the way "learn to code" was 10 years ago?
I am looking to pivot in my career. I have been unemployed several months and, for reasons I don't understand (market saturation?), no one is responding to my seemingly perfectly qualified applications to roles in line with my previous job.
OpenAI just put out a blog post which includes releasing an email chain involving Elon Musk, Sam Altman, and Ilya Sutskever, regarding Scott's SSC post "Should AI be Open?" https://openai.com/blog/openai-elon-musk
Someone sent the post to Musk, who forwarded it to Altman, Sutskever, and Brockman. Sutskever's reply:
> The article is concerned with a hard takeoff scenario: if a hard takeoff occurs, and a safe AI is harder to build than an unsafe one, then by opensorucing everything, we make it easy for someone unscrupulous with access to overwhelming amount of hardware to build an unsafe AI, which will experience a hard takeoff.
> As we get closer to building AI, it will make sense to start being less open. The Open in openAI means that everyone should benefit from the fruits of AI after its built, but it's totally OK to not share the science (even though sharing everything is definitely the right strategy in the short and possibly medium term for recruitment purposes).
To which Musk posted "Yup". Musk's agreement with Sutskever's rebuttal to Scott's article is now a major point in the lawsuit, which I find interesting.
Would removing restrictive zoning temporarily raise the cost of buying land or a house? My intuition is that by removing restrictive zoning, you are increasing the set of things that can be done with the land, including options that could be more profitable for the land owner than, say, a single family home. I would think that this should raise the price of buying land/a house, at least until increased construction raises the supply and drives down the price.
Curious as to peoples’ thoughts on this paper about electoral competition and electoral systems. Would a Condorcet-consistent electoral system solve the problem of hyper-polarized politics? Is it realistic to get this implemented?
Say what you will about journalism, but this is pretty scummy. It also seems weird that they're doing this *now* rather than back when Wordle was actually popular.
This is a thought-provoking and profound piece that touches on some big questions about human civilization, values, technology and the future. I’ll do my best to provide a high-level summary of the key points, but I would encourage you to read the full article if you have a chance, as it contains a lot of nuance and insight that is hard to capture in a short summary.
The central idea is that there are powerful forces and incentives that push civilization towards bad outcomes, even when no individual wants those outcomes. The author frames this in terms of a “Moloch” figure that demands sacrifice and bad behavior.
Some key examples of these “multipolar traps” that are discussed:
* Arms races between nations
* Competitive dynamics between firms that lead to poor treatment of workers and damage to the environment
* Evolved biological incentives that no longer serve our values
* Ideological and political races to the bottom
The concerning thing is that as technology grows more powerful, the potential danger of these traps increases. Advanced technologies could lead to terrible outcomes if not very carefully managed.
The author suggests the only way out is to create a singleton superintelligent AI that is imbued with human values and has the power to prevent these destructive forces. This is an ambitious but perilous proposal.
The article covers these ideas in much more depth, with vivid examples and poetic language. While heavy and sobering, it is a rewarding read for those interested in the trajectory of human civilization. The ideas are speculative but the core concerns seem important to grapple with. I’d recommend reading it in full if these topics interest you and you can find the time.
I also asked Claude to give me a description of Moloch for Midjourney and got:
> A towering, menacing bronze idol with the head of a bull, seated on a throne of gears, industrial machinery, and weapons. The idol has fiery eyes and smoke pouring from its nostrils. Thousands of tiny human figures are gathered around the base of the throne, some worshipping the idol, others being crushed by the gears or cast into flames. The scene is set against a bleak, dystopian cityscape with smokestacks and brutalist buildings looming in the background under an ominous red sky.
Entirely uncontroversial author Abigail Shrier is doing the promo rounds again, talking about her new book, Bad Therapy.
The main thesis being that we (as a society) are promoting and causing mental health issues in our kids by being too proactive in our approach. Putting too much focus on mental health, having under-qualified teachers, social workers and school counselors frequently screen them or ask them about mental health, preemptively putting kids in therapy when something sad happens, pathologizing and medicalizing the experience of being a teenager, etc. etc. has iatrogenic effects. In many cases it does the exact opposite of what it’s intended to do: making kids more depressed, fragile, and traumatized.
It all sounds very commonsensical but also pop-sciency. On one hand it meshes well with ideas and observations put forward by people like Jonathan Haidt and Jean Twenge, and with stats and trends like the one about antidepressants shared by Beowulf888 further down here. On the other hand, I don’t expect everyone to agree. If they did, I don’t see how it would be a growing problem.
So, does anyone here have perspectives on how seriously a parent should take this?
Tell me again how AI is so bloody wonderful, I dare you.
Another Irish company has decided that AI is the way to go for customer service. I'm currently tussling with it because it cannot understand my query. It is plainly running off a script and is even stupider than an ordinary human. It redirects me to the website, I've been on the website and the solution to my query is not there. It's not a technical issue or fault. The thing cannot understand English or think or do anything that an ordinary, average human could understand in five seconds.
Bah, humbug. And this is what is going to rule the sevagram?
Antidepressant usage among younger age cohorts is way up. I'll let the Results and Conclusions of this paper speak for themselves...
"RESULTS
Between January 2016 and December 2022, the monthly antidepressant dispensing rate increased 66.3%, from 2575.9 to 4284.8. Before March 2020, this rate increased by 17.0 per month (95% confidence interval: 15.2 to 18.8). The COVID-19 outbreak was not associated with a level change but was associated with a slope increase of 10.8 per month (95% confidence interval: 4.9 to 16.7). The monthly antidepressant dispensing rate increased 63.5% faster from March 2020 onwards compared with beforehand. In subgroup analyses, this rate increased 129.6% and 56.5% faster from March 2020 onwards compared with beforehand among females aged 12 to 17 years and 18 to 25 years, respectively. In contrast, the outbreak was associated with a level decrease among males aged 12 to 17 years and was not associated with a level or slope change among males aged 18 to 25 years."
"CONCLUSIONS
Antidepressant dispensing to adolescents and young adults was rising before the COVID-19 outbreak and rose 63.5% faster afterward. This change was driven by increased antidepressant dispensing to females and occurred despite decreased dispensing to male adolescents."
Earlier down thread I engaged with a user named FLWAB, regardless of what's the actual point we were debating, I behaved like an asshole. Some of this is due to (as perceived by me) unfairness on FLWAB's part, and some of it is emotional volatility on my part because the topic we were discussing relates to morality, and is important to me, and it's all too easy to appear heartless and cruel by speaking a certain way, and I perceived FLAWB to be heartless and cruel. FLWAB also attacked my objectivity and accused me of manipulation in what I perceived to be a personal attack.
This comment is:
(1) A public apology to FLWAB: I'm sorry I was an asshole. And I'm sorry about a more dismissive apology that I made in the comments that didn't properly acknowledge me being an asshole in the comments before it.
(2) A pledge to be more careful next time
(3) A promise conditional on FLWAB accepting my Apology, I will refrain from commenting for a number of Open Threads they choose, not exceeding 7 Open Threads
(4) A promise conditional on FLWAB accepting my Apology, I will watch/read a source sympathetic to their views of their choosing provided it doesn't exceed 100 page or 150 minutes
(5) A promise conditional on FLWAB accepting my Apology, I will delete my comments down thread if they so choose
So after watching Epic Rap Battle's "Ivan the Terrible vs Alexander the Great", it struck me that modern-day politicians might get tagged with titles like "The Great" or "The Terrible" in the history books. Does anybody have any guesses as to who'll get what?
I am still on my journey exploring Aldous Huxley's writing, this week slowly reading my way through Heaven and Hell. I was very surprised to find out he articulated the idea that our minds are inhabited by some quasi-independant entities, which we should catalogue and study "the way a zoologist studies animals found on a new continent". I wrote a short essay on this (https://valentinsocial.substack.com/p/aldous-huxley-on-the-exploration). I keep getting fascinated by his ability to write in a way which makes you feel like you are being told some galaxy-brain concepts, even when it's something super simple, like the fact that colors are more vivid while on mescaline
Regarding the stereotype that the Dutch are direct and the British highly indirect: if there is truth to it, what are the historical reasons why it is supposed to have evolved?
Scott has written a few times about Land Value Tax, the benefits of which seem overwhelming. Last year, CA Assembly member Alex Lee introduced a bill to study the potential impact of a LVT in California. That's a good sign, but even in an optimistic case it seems like adoption would be lightyears away in spite of the glowing cost-benefit profile.
Has anyone discussed the idea of getting a California proposition on the ballot to implement LVT? According to Ballotpedia, such a proposition was on the ballot over 100 years ago -- in 1918 -- and was crushed by 50 points. Getting a proposition on the ballot is no easy feat (I think it requires something like 500K signatures IIRC?) but could be an interesting angle if it sputters in the legislature.
> The Alameda County Democratic Central Committee is a deep down ballot race. They are large, misunderstood contests that often include virtual unknowns vying for the right to volunteer the first Wednesday night of every month to the Democratic cause.
> Understanding the Alameda County Democratic Central Committee might be easier if you just view them as the Alameda County Democratic Party, one of 58 franchises within the California Democratic Party.
> What is their role? To fundraise and support their endorsed candidates. In years past, the fundraising aspect was a weak point, but unexploited simply because Alameda County has one-party rule. But within the party there has been internal threats to the progressive majority on the central committee by business-friendly moderates in the Tri-Valley and Fremont.
> Compounding the schism was Pamela Price and her push to win control of the central committee as far back as her first run for Alameda County district attorney in 2018. The plan worked. In 2020, Price spent heavily to help herself and a slate of candidates win seats on the central committee.
...
> To underscore Price’s control of the central committee, the chaos followed an attempt to allocate up to $10,000 to the campaign opposing Price’s recall. As a central committee member, Price voted for the allocation.
> The central committee election this March is essentially about younger progressives moving the local party further left and experienced elected officials aiming for a return to the status quo.
I have been posting on Substack a subtle but I think important distinction about how we should be thinking about central bank macroeconomic management. The difference is that instead of thinking about managing aggregate supply and demand (implicitly as if the economy were one good produce with one input -- undifferentiated labor -- and therefor only one relative price, the real wage) we should think about the central bank as facilitating adjustments in relative prices among goods and differentiated labor in response to shocks that affect sectors differently. The "facilitation" is critical because all prices do not adjust at the same speed and some are downwardly sticky. This has implications for the average target rate of inflation that a central bank sets to facilitate "Brownian movement shocks" as well as how "flexible" it should be in engineering over target inflation the face of extraordinary shocks requiring extraordinary movements in relative prices.
I am seeking comments on these ideas, ideally on the Substack posts themselves or here if anyone is so moved.
Here are my questions for Scott regarding parenthood.
• What are your thoughts about school nowadays? Have they changed since a couple years ago?
• Have you read *Selfish Reasons to Have More Kids* and which parts do you agree with or disagree with?
• How are you going to try to affect which influences they end up under? Presumably you're going to limit and supervise device usage during the younger years, and then eventually give up and hope for the best.
• Are you going to make a big budget for individual tutoring? That was one of the things that stood out to me about the childhoods of Marcus Aurelius and John von Neumann, and which seems underrated.
• Are you going to prioritize social parallax, and if so, how? (In the sense of moving between multiple social groups, so as to see them all from different perspectives.)
Several years ago, I read an article claiming that the 5G technology used in the US broadcasts on the same frequency as the Doppler radar used for weather prediction, and that this would make forecasts less accurate [something something] climate change [something something] Donald Trump. Did anything come of this, or was it entirely made up?
Question - does anyone know what foods are traditional for St. Patrick's Day in Ireland? Especially finger foods? My mother and I are doing coffee hour for our church that Sunday, and wanted to do something a bit more than dying everything green.
Hi folks, I was wondering if someone might have advice on this purchase. What are best practices in buying a used car?
How do I find out if the price quoted for a used car by a seller is a good price?
I went to carfax.com and checked out reports of accidents or damage (whatever involved a police report will show up), and how many owners there were. No accidents, one owner only. Looks good.
It is a Toyota Camry.
I don't know a mechanic in the city to take it to, to get it checked out.
Do you think it's worth doing IVF to select for genes to potentially increase IQ at this point in time? Or is the risk/cost still not worth the benefit? Do any companies even offer this sort of thing yet?
There's been some good articles about envy on Substack. Rob Henderson had a good review of the book "Sadly, Porn" a while ago. But when I re-watched the older version of "The Razor's Edge" an adaptation of Somerset Maugham's novel, I was blown away by its portrayal of envy. My review here: https://falsechoices.substack.com/p/old-stories-the-razors-edge BTW, it's free to watch on YT.
What percent of income is OK to pay for the substack subscription? I feel like $10 is kinda too much for me to pay each month, but like $2.5 is too little. (I'm not poor, but that's by Russian standarts, plus we have some problems with money transfers).
AI Go programs have significantly improved the play of top Go players — not only because they are learning from them, but players are suddenly inventing new moves at a faster rate too! But this may be related to open-source Go programs like LeelaZero, though, which allow players to look at the decision-making process behind the moves.
Your question presumes libertarians want to "take down" some function of government.
For some, libertarianism is about understanding. One can't begin to understand until he stops looking for things to "take down."
A mob of over a hundred young people vandalized two local banks in our city last week. They claimed it was a protest against The Police and/or Palestine. (So many things to "take down;" so little time.)
None of the three local TV station covered it. One had to look in the local section of the one remaining newspaper to even know it occurred.
I'm curious as to how the general body of ACXers would approach the very contingent, very abstract, very difficult-to-measure (but ambiently very real-seeming) phenomenon of the competency crisis.
In just about any of the well-known crises of our day which you'd care to credit, there will lay at least some grounds for suggesting that the crisis in question is underpinned by a broader crisis in which we are losing the ability to produce competent people, appoint them to important posts, or both.
Bandied about a fair deal in the last bits of 2023 by Palladium magazine and, I believe, Edward Luttwak, but otherwise yet to gain much critical mass (probably because it directly implicates lots of institutions, like the press and civil services, that critical mass-drivers tend to like or be employed by).
Thanks for announcing Manifest tickets! But a correction on pricing: we originally said GA tickets were going to be $449 (not 499), but have lowered that to $399. And early bird tickets are $249 (only 10 left!).
What did you find interesting about Jessica Ocean's analysis? I found it vague to the point of uselessness: "laws can be implemented badly" is an all-purpose argument that doesn't answer the question of whether the new law will be implemented more or less badly than the previous law, and "more money doesn't always solve the problem" doesn't answer the question of whether this reallocation of existing money from one stream to another will make particular problems better or worse. As far as I can tell, the only meaningful criticism raised in that post which depended on the actual details of Prop 1 was the note that too much of the housing was set aside for veterans.
I’m confused about fractional reserve banking. The consensus seems to be that banks create money (see for example https://www.youtube.com/watch?v=O5DaC1Ujrrg), but I think that’s an oversimplification.
Take for example Capitol One, based on their annual report, as of year end they had $348 billion of deposits and $305 billion of loans outstanding. So roughly 90% of deposited were loaned out. The theory is all of those loans were deposited into a bank, so that 90% of those would be loaned out (and then deposited and then loaned out etc). But that can’t be right. If I charge $1,000 on my Capital One card, the money goes to the vendor who just replaces his inventory, he doesn’t deposit the money. Maybe just to the extent of his gross margin, maybe 10% at most.
The supplier receives payment from the vendor but they have to build another widget with the money, so they don’t deposit the full amount, they spend it on inputs to build another one. Maybe they keep 10% at most? And that’s assuming pretty high margins. If I spent $1,000 at restaurants, then they’re keeping maybe 2-3%.
Where have I gotten this wrong? The idea that $1 of deposits create 10x as much money through FRB seems hyperbolic.
I had put this on the February links post but got there very late, and likely no one will see it, so will re-post here. I'm surprised at how absolutely no one has commented on link 18
(adding only two links manually because substack is impossible)
Piketty and his diatribe on inequality have been at the forefront of the left crusade on capitalism and classical liberal style economics. And no one here finds the fact that Piketty et al were substantially incorrect worth comment?
I’ve been using Yahoo Finance for a long time to keep tabs on the US stock market and get quick quotes. Even as the whole Yahoo thing was sliding into irrelevant mess of crap, with the help of adblockers the Finance page was useable.
Not anymore, with the new design it’s truly jumped the shark.
So: anybody has a recommendation for a clean-looking site for this?
For some inexplicable reason, GOOGLE is killing Google Podcasts on 2 April 2024. That's soon!
I subscribe to about a million different free podcasts. Where should I get them now? GOOGLE wants me to get them from YouTube Music but I'd rather not since they killed my prior favorite Stitcher and now are actively homiciding Google Podcasts.
Where do I go? Spotify? Apple Podcasts? Some other place?
Does anyone here have any memories or articles about how the Western left - the real left, I mean - reacted to the collapse of the Soviet Union? I'm finding it weirdly hard to find information about it, and I'm interested.
I like to stay clear of culture war topics, and what once was called the IWD and is now probably more clustered as the heterodox skeptics community has been too close to it for me. Coleman Hughes used to be an exception for me, but this excellent 3 part series made me update on him as well: https://radleybalko.substack.com/p/the-retconning-of-george-floyd
Is 2024 a good year to visit Israel? Due to the war, I imagine tourism is down, so there won't be crowds at the main sites, and plane tickets and hotels will be cheaper.
If any of you live in Israel or frequently visit, can you give me advice?
I write about specific high-impact ways I've improved my life. N=1, but most of them feel pretty generalizable. Every other post on philosophy or current events so you don't feel like you're being lectured.
Back in 1997 David Shumway published “The Star System in Literary Studies” in PMLA. He begins with a paragraph about George Lyman Kittredge, of Harvard’s English Department at the end of the 19th and beginning of the 20th century, noting that Kittredge was unknown to the public. Here’s the first sentence of his second paragraph:
"Kittredge, who virtually founded Chaucer studies in the United States, stood at the head of a professional genealogy that controlled the field for many years after his death, but he was not a star. Nor were any of his illustrious contemporaries or near contemporaries, such as John Manly, John Livingston Lowes, and so on. Why they were not stars and Judith Butler, Jacques Derrida, Stanley Fish, Henry Louis Gates, Jr., Fredric Jameson, Gayatri Chakravorty Spivak, and other figures in the academy are is the subject of this essay."
What I’m wondering is whether or not the so-called AI “Godfathers” don’t represent a similar phenomenon in contemporary AI. Strictly speaking I believe the Godfather term applies to the three winners of the 2018 Turing Award, Yoshua Bengio, Yann Lecun, and Geoffrey Hinton, but I believe there are others in AI with a similar status, such as Ilya Sutskever, Hinton’s student and co-founder of OpenAI, Andrej Karpathy, the former director of AI at Tesla who just made waves, albeit little ones, by resigning from OpenAI, Demis Hassibis, cofounder of DeepMind, and perhaps even such figures as Nick Bostrom and Eliezer Yudkowsky, who aren’t AI researchers but are highly influential figures through their commentary. Perhaps Sam Altman, the heroic CEO who fought off a recalcitrant board, is a star as well.
Later:
Concerning the rise of literary stars, Shumway noted: “As theory has called into question the traditional means by which knowledge has been authorized, it may be that the construction of the individual personality has become an epistemological necessity.” That seems like the state of AI today. We’ve got a very complicated technology involving a blend of engineering, science, and alchemy, lacking objective knowledge. Note only that, the technology is enormously important and will change the way we live. In the absence of objective knowledge, what choice do we have but to steer by the freakin' stars?
Question for those knowledgeable about the USSR's international relations/global left:
My rough understanding is that following the dissolution of the USSR, with the exception of Cuba and North Korea, most nations/political leaders aligned with the USSR ultimately pursued their own paths. In today's context, there seems to be little that unites these countries.
However, nearly all the nations/political leaders who still broadly adhere to their far-left/USSR roots (e.g., Lula, Daniel Ortega, the ANC, etc.) continue to hold a strong animosity towards Israel, seemingly as much as they did during the USSR era. This also applies to every day Westerners who were sympathetic to the USSR.
I'm seeking a deeper understanding of this phenomenon. These nations/political leaders/people appear less focused on the USSR/far-leftism nowadays and generally exhibit significant differences in their contemporary ideologies, which makes me believe that any positions on this aren't rooted in some underlying ideology. Moreover, Israel does not seem particularly relevant to their countries to have needed to maintain a fixed position on the subject. Therefore, I'm trying to comprehend why opposition to Israel has remained so pronounced and become entrenched in this group. Is the opposition to Israel rooted in some ideology that transcends socialism...? (I'm skeptical) or is Israel merely being used as a proxy for antipathy towards the United States, which would make much more sense as something that would remain sticky? Or some third thing...
I'm curious to hear people's thoughts on this... especially if there are other similar things like opposition to Israel that the post-USSR players have remained unified on.
Scott wrote a great article on IRB's in science in 2014. Universities go way beyond what the the federal government requires, which is silly enough. And universities take 50% of grants as overhead. Are there any universities which do science like Hillsdale does humanities, intelligently with minimum administration adn overhead? Could somebody start one? I bet it would be easier to lure top scientists away with lack of bureaucracy than with high salaries.
Do we know when we're getting the resolutions for the 2023 predictions? Specifically for the ones who are still open on Metaculus (https://www.metaculus.com/tournament/2023-contest/). It's been already two months and if another month and a bit passes they will count for another leaderboard, which is not ideal.
> We are concerned about late-stage AGI development becoming a competitive race without time for adequate safety precautions. Therefore, if a value-aligned, safety-conscious project comes close to building AGI before we do, we commit to stop competing with and start assisting this project. We will work out specifics in case-by-case agreements, but a typical triggering condition might be “a better-than-even chance of success in the next two years.”
Scott then asked to set some reminders to check back in March 2024 whether these plans for specific triggering conditions of the merge-and-assist clause went anywhere.
As far as I can tell after some internet search, there was no public announcement of such firmer commitments by OpenAI. There have been several further statements on their safety plans, including the Superalignment Team, Frontier Model Forum and Voluntary AI commitments (July 2023), but no improvements on the merge-and-assist mechanism (see their blog
In fact, in December 2023 the company VERSES (whose Chief Science Officer is Karl Friston) published an open letter in which they announced having a pathway to AGI and asking OpenAI to trigger the above clause in their charter:
As I prepare to not vote in the next Irish referendum on something stupid in the constitution being replaced by something even stupider, I am confirmed in my bias - unique to myself in this country - that constitutions are largely useless pieces of paper and you can organise a society perfectly well without one.
For all its faults, and despite being the home of my eternal racial enemies, I have to admit that the U.K. gets on perfectly fine without a formal constitution and remains a relatively free and relatively prosperous country.
Meanwhile North Korea has and the Soviet Union had constitutions, the Soviet one guaranteeing absolute freedom of speech, decide amongst yourself if that worked or not.
Constitutions are either the dictatorship of the past upon the present subject to the whims of unelected justices - often leading to an effective krytocracy - or worse, the legislature uses referenda to not govern, or legislate, and which without getting into the specifics in Ireland that’s the case with the specific amendment here.
I feel the same about initiatives in US States as well. I’ve elected you to govern - don’t ask me to legislate.
So is it time for one last referendum or constitutional convention or whatever is needed in your country - to get rid of the constitution?
There are maybe three or four philosophers per century who make a difference in the world. Meanwhile, there are thousands and thousands of philosophers churning out philosophy papers.
What are those papers for? Who reads them? What are philosophers for?
Current implied probabilities for US presidential election, Betfair/Polymarket/Metaculus/Manifold (percentages, nearest whole number):
Trump: 47/54/50/48
Biden: 28/32/45/47
Obama: 8/5/1/1
Newsom: 4/3/1/0
Harris: 3/2/2/1
Haley: 3/1/1/0
Kennedy: 3/3/1/1
Since I last posted about this, the gap between Betfair and Polymarket on Trump has narrowed slightly, which is good to see, but the difference is still surprisingly large. One person now holds 28% of the Trump "Yes" shares on Polymarket, which means they have now bet over $1m. Biden continues to be much lower in real money markets than other prediction sites; one side must be wrong.
Betfair continues (IMO) to overvalue the outsider candidates, which probably represents small-odds bias, i.e. the difficulty of taking large "no" positions at long odds on an event 8-10 months in the future. This appears true to a lesser extent for Polymarket, which is probably because their market claims "supports negative risk (convert No shares to Yes of the other options)". For example, only 311 "no" shares seem to exist for Harris, despite there being 2,968,743 "yes" shares. Assuming this has actually been implemented correctly, it's a clever idea. It's worth noting that this means the bet totals are inaccurate: IIUC much less than $3m has been bet on Harris for example.
Weird question: a while ago, I went to see a therapist because my executive function was/is terrible and I suspected ADHD. After the therapist had dug into it, it appeared that actually I don’t have ADHD, just ASD, which also involve some executive dysfunction (which is probably made worse by eg. my low self-confidence, etc., I guess). So, I have ASD, I’m fine with that, but now my therapist says "maybe you should consider getting ADHD stimulants, those might help with your executive function".
She’s probably right, at least it makes some kind of sense, but… I can’t just go to a psychiatrist, diagnosis in hand, saying "here it says I don’t have ADHD, now gimme the ADHD drugs anyway, pretty please?", and expect it to work! Or can I? Even assuming that I can: I’ve never seen a comparison of the various kinds of stimulants there are, from Adderall to weird nootropics to very very strong coffee. Anyone has good sources on what works best / what has the best "effects on executive function : difficulty of getting it and magnitude of side effects" ratio?
Is the age of formal classroom schooling entering a long secular decline? And if so - apart from the growth of homeschooling - what could possibly replace it as the primary means by which society teaches its children? https://grahamcunningham.substack.com/p/teach-your-children-well
There isn’t much literature on this, so I’m asking here: if you were on 150mg Wellbutrin and 60 mg buspirone, would you avoid nonalcoholic beer (defined as less than 0.5% ABV)?
All literature agrees you should avoid alcohol on these medications. Studies haven’t really been done on NA beers like Athletic Brewery which are increasingly popular. Any thoughts, inferences, or guidance is appreciated.
I used to read the Economist, but lately I have less time and can only get to "Politics this week". I like it a lot, but on its own it's not worth the subscription. Does anybody know of a good replacement? Something to tell me the important news from the recent week around the world, in very brief form?
Proposition 1 adds additional criminal penalties to homelessness and nothing justifies that. The state has not shown a willingness to abide by restrictions it places on itself, so a promise to provide services in exchange for making certain things double illegal is really just a promise to make things double illegal.
Can someone who knows the science please explain to me: Does IQ matter?
Many years ago I read Stephen Gouldings The Mismeasure of Man, where he says there have been many attempts to measure human intelligence on a straight line, and they have all been flawed. And that IQ is just one of the latest of these. He said IQ had been created to separate children needing special attention from other kids, and was not meant to be used for individuals.
But in recent years, there have been many people claiming IQ is important. The chief proponent I remember is Jordan Peterson who claims IQ is the main indicator of how much you will earn, and if your IQ is less than 80 even the US Army won't hire you. This 2nd claim got some criticism online and some people said it was a made up fact.
But I dont know enough about this field--how relevant is IQ? Is it really as important as many people are claiming it to be?
My latest paper about ChatGPT: Does ChatGPT Understand the Concept of Tragedy? Symbolic AI & Neural Nets
Abstract: Tragedy is an abstract concept. David Hays proposed that abstract concepts can be metalingually defined. A term, such as tragedy, can be given its meaning by a pattern over a string of words. By using a cognitive network to define tragedy Brian Phillips investigated that computationally. In this paper I have taken Brian Phillips’ definition and presented it to ChatGPT in six different trials, where each trial also included a short story. Three stories satisfied the definition and three did not. ChatGPT was asked to indicate whether or not the story satisfied the definition it was given. It was correct in each case.
https://new-savanna.blogspot.com/2024/03/does-chatgpt-understand-concept-of.html
Anyone else find it weird how OpenAI goes to frankly absurd lengths to be "unbiased" (including flat out lying about certain sensitive topics), presumably to avoid pissing off the wrong people and getting negative attention for OpenAI and its services. But then, the CEO of OpenAI has as his twitter bio a symbol of his ethnoreligious identity, which he set this as in the midst of one of the bloodiest and brutal ethnoreligious conflicts in modern history, one which is amongst the most sensitive political topics in the world right now. Really gets the noggin joggin.
Has anyone else noticed Google Doc's autocorrect get much more aggressive and stupider recently?
Previously, it would just attempt to correct mispellings or grammar errors. Sometimes it would get one wrong, but for the most part it was ok, and it would never attempt to apply the same correction more than once if you backspaced and changed it back. However, perhaps a few weeks ago, it got a lot more aggressive and also stupider, and now frequently makes changes that aren't even gramatically correct.
For example, when I was writing just now, in the space of one paragraph, it changed the "worn" to "worms" in "However, there were also small holes worn", and it changed "I applied" to "I apple".
Furthermore, it no longer gives up if you undo its correction. I've seen it apply the exact same (mis)correction three or more times when I backspace and fix the "correction".
Does anyone here know what's going on here? I'm curious what people on Google Docs would say about this.
Oh, Jebus, how ‘bout that Caitlin Clark?
https://www.foxnews.com/sports/iowa-womens-basketball-storms-back-beat-nebraska-win-big-ten-tournament-slow-caitlin-clark-start.amp
He's an update on the kid-prompted, generative-AI-generated, smart-phone distributed kiddie porn saga at Beverly Hills middle school. Five of the 16 perps (err students )involved have been expelled under a stipulated agreement with the parents of the perps (so the parents have accepted the expulsion and they won't legally challenge the school district). It's unclear what if anything is going to happen to the other 11 perps (err students) involved.
The county DA and BHPD are still investigating, so criminal charges could still be filed. But... "California’s laws against possessing child pornography and sharing nonconsensual nude pictures do not specifically apply to AI-generated images, which legal experts say would pose a problem for prosecutors."
It's unclear if the parents of the victims of this AI-generated classmate porn will sue the parents of the perps or the school district.
https://www.latimes.com/california/story/2024-03-07/beverly-hills-school-district-expels-8th-graders-involved-in-fake-nude-scandal
OC ACXLW Sat March 9 AI vs The Past, Who is Utilitarian?
Hello Folks!
We are excited to announce the 58th Orange County ACX/LW meetup, happening this Saturday and most Saturdays after that.
Host: Michael Michalchik
Email: michaelmichalchik@gmail.com (For questions or requests)
Location: 1970 Port Laurent Place
(949) 375-2045
Date: Saturday, Mar 9 2024
Time 2 pm
Conversation Starters :
AI’s Threat to History:
Text:
A.I. Is Coming for the Past, Too
https://docs.google.com/document/d/1pj2GL3s-R4uIy9udrSGcOMQCNUk-6dVcaWNFTh-mTmg/edit?usp=sharing
Audio:
AI's Threat to History - The AI Breakdown: Daily Artificial Intelligence News and Discussions (podcast) | Listen Notes
https://www.listennotes.com/podcasts/the-ai-breakdown/ais-threat-to-history-Kw2Jm8CrBkc/
Less Utilitarian than Thou by Scott Alexander:
https://www.astralcodexten.com/p/less-utilitarian-than-thou
Thinking the unthinkable by Tetlock:
Thinking the unthinkable tetlock2003.pdf
https://drive.google.com/file/d/1RAXxXggyCSYoK9aUI4nZ3MdPW9s9wrE3/view?usp=sharing
Walk & Talk: We usually have an hour-long walk and talk after the meeting starts. Two mini-malls with hot takeout food are readily accessible nearby. Search for Gelson's or Pavilions in the zip code 92660.
Share a Surprise: Tell the group about something unexpected that changed your perspective on the universe.
Future Direction Ideas: Contribute ideas for the group's future direction, including topics, meeting types, activities, etc.
I am a somewhat intelligent person -- I once measured with a 143 IQ, which seems a fair number to capture how I qualitatively think of my intelligence. Is there any fast path to a decently paying job the way "learn to code" was 10 years ago?
I am looking to pivot in my career. I have been unemployed several months and, for reasons I don't understand (market saturation?), no one is responding to my seemingly perfectly qualified applications to roles in line with my previous job.
OpenAI just put out a blog post which includes releasing an email chain involving Elon Musk, Sam Altman, and Ilya Sutskever, regarding Scott's SSC post "Should AI be Open?" https://openai.com/blog/openai-elon-musk
Someone sent the post to Musk, who forwarded it to Altman, Sutskever, and Brockman. Sutskever's reply:
> The article is concerned with a hard takeoff scenario: if a hard takeoff occurs, and a safe AI is harder to build than an unsafe one, then by opensorucing everything, we make it easy for someone unscrupulous with access to overwhelming amount of hardware to build an unsafe AI, which will experience a hard takeoff.
> As we get closer to building AI, it will make sense to start being less open. The Open in openAI means that everyone should benefit from the fruits of AI after its built, but it's totally OK to not share the science (even though sharing everything is definitely the right strategy in the short and possibly medium term for recruitment purposes).
To which Musk posted "Yup". Musk's agreement with Sutskever's rebuttal to Scott's article is now a major point in the lawsuit, which I find interesting.
Would removing restrictive zoning temporarily raise the cost of buying land or a house? My intuition is that by removing restrictive zoning, you are increasing the set of things that can be done with the land, including options that could be more profitable for the land owner than, say, a single family home. I would think that this should raise the price of buying land/a house, at least until increased construction raises the supply and drives down the price.
I'm so late to this party, but nonetheless popping in to plug my latest on collectives, individuals, and Boyz II Men. For your enjoyment.
https://kyleimes.substack.com/p/there-is-a-cycle-that-never-stops
Curious as to peoples’ thoughts on this paper about electoral competition and electoral systems. Would a Condorcet-consistent electoral system solve the problem of hyper-polarized politics? Is it realistic to get this implemented?
Link to a blog post describing the paper: https://edwardbfoley.substack.com/p/a-must-read-analysis-of-why-us-elections
Link to paper: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4728225
Apparently the NYT is now abusing DMCA to take down all Wordle-like games (https://github.com/cwackerfuss/reactle/issues/765)
Say what you will about journalism, but this is pretty scummy. It also seems weird that they're doing this *now* rather than back when Wordle was actually popular.
Sorry to hijack your post, but I couldn't post on your old site.
I used Claude (AI) to summarise your Meditations on Moloch, but also to create some associated imagery, which I thought you might be interested in.
https://www.kublermdk.com/2024/03/06/meditations-on-moloch-summary/
-----------
https://slatestarcodex.com/2014/07/30/meditations-on-moloch/
This is a thought-provoking and profound piece that touches on some big questions about human civilization, values, technology and the future. I’ll do my best to provide a high-level summary of the key points, but I would encourage you to read the full article if you have a chance, as it contains a lot of nuance and insight that is hard to capture in a short summary.
The central idea is that there are powerful forces and incentives that push civilization towards bad outcomes, even when no individual wants those outcomes. The author frames this in terms of a “Moloch” figure that demands sacrifice and bad behavior.
Some key examples of these “multipolar traps” that are discussed:
* Arms races between nations
* Competitive dynamics between firms that lead to poor treatment of workers and damage to the environment
* Evolved biological incentives that no longer serve our values
* Ideological and political races to the bottom
The concerning thing is that as technology grows more powerful, the potential danger of these traps increases. Advanced technologies could lead to terrible outcomes if not very carefully managed.
The author suggests the only way out is to create a singleton superintelligent AI that is imbued with human values and has the power to prevent these destructive forces. This is an ambitious but perilous proposal.
The article covers these ideas in much more depth, with vivid examples and poetic language. While heavy and sobering, it is a rewarding read for those interested in the trajectory of human civilization. The ideas are speculative but the core concerns seem important to grapple with. I’d recommend reading it in full if these topics interest you and you can find the time.
—
Side note: Scott Alexander is now posting to the Astral Codex Ten substack https://www.astralcodexten.com/
I also asked Claude to give me a description of Moloch for Midjourney and got:
> A towering, menacing bronze idol with the head of a bull, seated on a throne of gears, industrial machinery, and weapons. The idol has fiery eyes and smoke pouring from its nostrils. Thousands of tiny human figures are gathered around the base of the throne, some worshipping the idol, others being crushed by the gears or cast into flames. The scene is set against a bleak, dystopian cityscape with smokestacks and brutalist buildings looming in the background under an ominous red sky.
Every major company eventually becomes a defence contractor.
https://www.youtube.com/watch?v=6xRaPijFckA
Entirely uncontroversial author Abigail Shrier is doing the promo rounds again, talking about her new book, Bad Therapy.
The main thesis being that we (as a society) are promoting and causing mental health issues in our kids by being too proactive in our approach. Putting too much focus on mental health, having under-qualified teachers, social workers and school counselors frequently screen them or ask them about mental health, preemptively putting kids in therapy when something sad happens, pathologizing and medicalizing the experience of being a teenager, etc. etc. has iatrogenic effects. In many cases it does the exact opposite of what it’s intended to do: making kids more depressed, fragile, and traumatized.
It all sounds very commonsensical but also pop-sciency. On one hand it meshes well with ideas and observations put forward by people like Jonathan Haidt and Jean Twenge, and with stats and trends like the one about antidepressants shared by Beowulf888 further down here. On the other hand, I don’t expect everyone to agree. If they did, I don’t see how it would be a growing problem.
So, does anyone here have perspectives on how seriously a parent should take this?
Tell me again how AI is so bloody wonderful, I dare you.
Another Irish company has decided that AI is the way to go for customer service. I'm currently tussling with it because it cannot understand my query. It is plainly running off a script and is even stupider than an ordinary human. It redirects me to the website, I've been on the website and the solution to my query is not there. It's not a technical issue or fault. The thing cannot understand English or think or do anything that an ordinary, average human could understand in five seconds.
Bah, humbug. And this is what is going to rule the sevagram?
Antidepressant usage among younger age cohorts is way up. I'll let the Results and Conclusions of this paper speak for themselves...
"RESULTS
Between January 2016 and December 2022, the monthly antidepressant dispensing rate increased 66.3%, from 2575.9 to 4284.8. Before March 2020, this rate increased by 17.0 per month (95% confidence interval: 15.2 to 18.8). The COVID-19 outbreak was not associated with a level change but was associated with a slope increase of 10.8 per month (95% confidence interval: 4.9 to 16.7). The monthly antidepressant dispensing rate increased 63.5% faster from March 2020 onwards compared with beforehand. In subgroup analyses, this rate increased 129.6% and 56.5% faster from March 2020 onwards compared with beforehand among females aged 12 to 17 years and 18 to 25 years, respectively. In contrast, the outbreak was associated with a level decrease among males aged 12 to 17 years and was not associated with a level or slope change among males aged 18 to 25 years."
"CONCLUSIONS
Antidepressant dispensing to adolescents and young adults was rising before the COVID-19 outbreak and rose 63.5% faster afterward. This change was driven by increased antidepressant dispensing to females and occurred despite decreased dispensing to male adolescents."
https://publications.aap.org/pediatrics/article/153/3/e2023064245/196655/Antidepressant-Dispensing-to-US-Adolescents-and?autologincheck=redirected
What do you think of Alzheimer's being caused by herpes? See this: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5935641/.
Earlier down thread I engaged with a user named FLWAB, regardless of what's the actual point we were debating, I behaved like an asshole. Some of this is due to (as perceived by me) unfairness on FLWAB's part, and some of it is emotional volatility on my part because the topic we were discussing relates to morality, and is important to me, and it's all too easy to appear heartless and cruel by speaking a certain way, and I perceived FLAWB to be heartless and cruel. FLWAB also attacked my objectivity and accused me of manipulation in what I perceived to be a personal attack.
This comment is:
(1) A public apology to FLWAB: I'm sorry I was an asshole. And I'm sorry about a more dismissive apology that I made in the comments that didn't properly acknowledge me being an asshole in the comments before it.
(2) A pledge to be more careful next time
(3) A promise conditional on FLWAB accepting my Apology, I will refrain from commenting for a number of Open Threads they choose, not exceeding 7 Open Threads
(4) A promise conditional on FLWAB accepting my Apology, I will watch/read a source sympathetic to their views of their choosing provided it doesn't exceed 100 page or 150 minutes
(5) A promise conditional on FLWAB accepting my Apology, I will delete my comments down thread if they so choose
So after watching Epic Rap Battle's "Ivan the Terrible vs Alexander the Great", it struck me that modern-day politicians might get tagged with titles like "The Great" or "The Terrible" in the history books. Does anybody have any guesses as to who'll get what?
I am still on my journey exploring Aldous Huxley's writing, this week slowly reading my way through Heaven and Hell. I was very surprised to find out he articulated the idea that our minds are inhabited by some quasi-independant entities, which we should catalogue and study "the way a zoologist studies animals found on a new continent". I wrote a short essay on this (https://valentinsocial.substack.com/p/aldous-huxley-on-the-exploration). I keep getting fascinated by his ability to write in a way which makes you feel like you are being told some galaxy-brain concepts, even when it's something super simple, like the fact that colors are more vivid while on mescaline
Regarding the stereotype that the Dutch are direct and the British highly indirect: if there is truth to it, what are the historical reasons why it is supposed to have evolved?
Scott has written a few times about Land Value Tax, the benefits of which seem overwhelming. Last year, CA Assembly member Alex Lee introduced a bill to study the potential impact of a LVT in California. That's a good sign, but even in an optimistic case it seems like adoption would be lightyears away in spite of the glowing cost-benefit profile.
Has anyone discussed the idea of getting a California proposition on the ballot to implement LVT? According to Ballotpedia, such a proposition was on the ballot over 100 years ago -- in 1918 -- and was crushed by 50 points. Getting a proposition on the ballot is no easy feat (I think it requires something like 500K signatures IIRC?) but could be an interesting angle if it sputters in the legislature.
For SF East Bay Democratic party voters:
I was trying to figure out what to do for "County Central Committee" and ran across this:
https://eastbayinsiders.substack.com/p/primarys-central-question
> The Alameda County Democratic Central Committee is a deep down ballot race. They are large, misunderstood contests that often include virtual unknowns vying for the right to volunteer the first Wednesday night of every month to the Democratic cause.
> Understanding the Alameda County Democratic Central Committee might be easier if you just view them as the Alameda County Democratic Party, one of 58 franchises within the California Democratic Party.
> What is their role? To fundraise and support their endorsed candidates. In years past, the fundraising aspect was a weak point, but unexploited simply because Alameda County has one-party rule. But within the party there has been internal threats to the progressive majority on the central committee by business-friendly moderates in the Tri-Valley and Fremont.
> Compounding the schism was Pamela Price and her push to win control of the central committee as far back as her first run for Alameda County district attorney in 2018. The plan worked. In 2020, Price spent heavily to help herself and a slate of candidates win seats on the central committee.
...
> To underscore Price’s control of the central committee, the chaos followed an attempt to allocate up to $10,000 to the campaign opposing Price’s recall. As a central committee member, Price voted for the allocation.
> The central committee election this March is essentially about younger progressives moving the local party further left and experienced elected officials aiming for a return to the status quo.
I have been posting on Substack a subtle but I think important distinction about how we should be thinking about central bank macroeconomic management. The difference is that instead of thinking about managing aggregate supply and demand (implicitly as if the economy were one good produce with one input -- undifferentiated labor -- and therefor only one relative price, the real wage) we should think about the central bank as facilitating adjustments in relative prices among goods and differentiated labor in response to shocks that affect sectors differently. The "facilitation" is critical because all prices do not adjust at the same speed and some are downwardly sticky. This has implications for the average target rate of inflation that a central bank sets to facilitate "Brownian movement shocks" as well as how "flexible" it should be in engineering over target inflation the face of extraordinary shocks requiring extraordinary movements in relative prices.
I am seeking comments on these ideas, ideally on the Substack posts themselves or here if anyone is so moved.
Some of the posts are:
https://thomaslhutcheson.substack.com/p/framework-for-monetary-policy-1
https://thomaslhutcheson.substack.com/p/arrrrr
https://thomaslhutcheson.substack.com/p/supply-and-demand-in-disinflation
https://thomaslhutcheson.substack.com/p/why-no-recession
https://thomaslhutcheson.substack.com/p/the-lessons-of-pandemic-inflation
Thanks
Here are my questions for Scott regarding parenthood.
• What are your thoughts about school nowadays? Have they changed since a couple years ago?
• Have you read *Selfish Reasons to Have More Kids* and which parts do you agree with or disagree with?
• How are you going to try to affect which influences they end up under? Presumably you're going to limit and supervise device usage during the younger years, and then eventually give up and hope for the best.
• Are you going to make a big budget for individual tutoring? That was one of the things that stood out to me about the childhoods of Marcus Aurelius and John von Neumann, and which seems underrated.
• Are you going to prioritize social parallax, and if so, how? (In the sense of moving between multiple social groups, so as to see them all from different perspectives.)
• A while ago I jotted 70 bullet points about raising children ( https://www.lesswrong.com/posts/6vPvpTZTBqe6evmKL/some-random-parenting-ideas ). I would be interested to hear if any of them stand out to you as good/bad.
Several years ago, I read an article claiming that the 5G technology used in the US broadcasts on the same frequency as the Doppler radar used for weather prediction, and that this would make forecasts less accurate [something something] climate change [something something] Donald Trump. Did anything come of this, or was it entirely made up?
Question - does anyone know what foods are traditional for St. Patrick's Day in Ireland? Especially finger foods? My mother and I are doing coffee hour for our church that Sunday, and wanted to do something a bit more than dying everything green.
Hi folks, I was wondering if someone might have advice on this purchase. What are best practices in buying a used car?
How do I find out if the price quoted for a used car by a seller is a good price?
I went to carfax.com and checked out reports of accidents or damage (whatever involved a police report will show up), and how many owners there were. No accidents, one owner only. Looks good.
It is a Toyota Camry.
I don't know a mechanic in the city to take it to, to get it checked out.
How do I figure out if the price is good?
So, how about Dune: Part 2?
Rot13: V gubhtug Nhfgva Ohgyre jnf n terng Srlq-Enhgun. Terng hfr bs zvpebrkcerffvbaf va znal fprarf, yvxr gur bar jurer Oneba Iynq trgf zhexrq.
Gurer jnf n ybg zber bs punatrf gb gur obbx guna va gur svefg zbivr, cnegvphyneyl Punav'f rkcnaqrq ebyr. V unq frra crbcyr pbzcynva nobhg guvf orsber gur zbivr naq unq qvfzvffrq vg nf fvzvyne gb fbzr crbcyr ovgpuvat nobhg Ynql Wrffvpn'f terngre ebyr va Cneg 1, ohg urer gur pbzcynvagf unq zber inyrapr - juvyr V'ir arire haqrefgbbq gur crbcyr jub guvax gung Mraqnln vf htyl, cnegvphyneyl fvapr fur'f cynlvat n fheiviny-bevragrq qrfreg abznq urer, ohg fur'f whfg abg gung tbbq bs na npgbe, naq V'z abg fher ubj Ivyyrarhir'f tbvat gb unaqyr Zrffvnu jvgu gur punatrf gurl qvq urer. BGBU n ybg bs phggvat qrpvfvbaf jrer tbbq (gur pbashfvat Thearl-if-Wrffvpn cybg sebz gur obbx, sbe vafgnapr, naq V jnf rira BX jvgu ubj Nyvn jnf unaqyrq, cebonoyl orggre gb qb vg yvxr Qravf qvq urer vafgrnq bs n zheqre gbqqyre.)
Do you think it's worth doing IVF to select for genes to potentially increase IQ at this point in time? Or is the risk/cost still not worth the benefit? Do any companies even offer this sort of thing yet?
There's been some good articles about envy on Substack. Rob Henderson had a good review of the book "Sadly, Porn" a while ago. But when I re-watched the older version of "The Razor's Edge" an adaptation of Somerset Maugham's novel, I was blown away by its portrayal of envy. My review here: https://falsechoices.substack.com/p/old-stories-the-razors-edge BTW, it's free to watch on YT.
What percent of income is OK to pay for the substack subscription? I feel like $10 is kinda too much for me to pay each month, but like $2.5 is too little. (I'm not poor, but that's by Russian standarts, plus we have some problems with money transfers).
Does anyone know of a place to donate to support the rollout of the new malaria vaccine?
Is "El Grillo" by Josquin des Prez the earliest (possibly-a-)joke song that we know of?
https://www.youtube.com/watch?v=OI-bQ0RkArA
If you do the voting guide even after Tuesday, it’d still be fun to read.
AI Go programs have significantly improved the play of top Go players — not only because they are learning from them, but players are suddenly inventing new moves at a faster rate too! But this may be related to open-source Go programs like LeelaZero, though, which allow players to look at the decision-making process behind the moves.
https://t.co/5Q6ZTiyPn1
Your question presumes libertarians want to "take down" some function of government.
For some, libertarianism is about understanding. One can't begin to understand until he stops looking for things to "take down."
A mob of over a hundred young people vandalized two local banks in our city last week. They claimed it was a protest against The Police and/or Palestine. (So many things to "take down;" so little time.)
None of the three local TV station covered it. One had to look in the local section of the one remaining newspaper to even know it occurred.
I'm curious as to how the general body of ACXers would approach the very contingent, very abstract, very difficult-to-measure (but ambiently very real-seeming) phenomenon of the competency crisis.
In just about any of the well-known crises of our day which you'd care to credit, there will lay at least some grounds for suggesting that the crisis in question is underpinned by a broader crisis in which we are losing the ability to produce competent people, appoint them to important posts, or both.
Bandied about a fair deal in the last bits of 2023 by Palladium magazine and, I believe, Edward Luttwak, but otherwise yet to gain much critical mass (probably because it directly implicates lots of institutions, like the press and civil services, that critical mass-drivers tend to like or be employed by).
Article length exploration here (also one for aficionados of the Rev./Napoleonic wars) https://heirtothethought.substack.com/p/the-competency-crisis-and-what-it
Thanks for announcing Manifest tickets! But a correction on pricing: we originally said GA tickets were going to be $449 (not 499), but have lowered that to $399. And early bird tickets are $249 (only 10 left!).
What did you find interesting about Jessica Ocean's analysis? I found it vague to the point of uselessness: "laws can be implemented badly" is an all-purpose argument that doesn't answer the question of whether the new law will be implemented more or less badly than the previous law, and "more money doesn't always solve the problem" doesn't answer the question of whether this reallocation of existing money from one stream to another will make particular problems better or worse. As far as I can tell, the only meaningful criticism raised in that post which depended on the actual details of Prop 1 was the note that too much of the housing was set aside for veterans.
I’m confused about fractional reserve banking. The consensus seems to be that banks create money (see for example https://www.youtube.com/watch?v=O5DaC1Ujrrg), but I think that’s an oversimplification.
Take for example Capitol One, based on their annual report, as of year end they had $348 billion of deposits and $305 billion of loans outstanding. So roughly 90% of deposited were loaned out. The theory is all of those loans were deposited into a bank, so that 90% of those would be loaned out (and then deposited and then loaned out etc). But that can’t be right. If I charge $1,000 on my Capital One card, the money goes to the vendor who just replaces his inventory, he doesn’t deposit the money. Maybe just to the extent of his gross margin, maybe 10% at most.
The supplier receives payment from the vendor but they have to build another widget with the money, so they don’t deposit the full amount, they spend it on inputs to build another one. Maybe they keep 10% at most? And that’s assuming pretty high margins. If I spent $1,000 at restaurants, then they’re keeping maybe 2-3%.
Where have I gotten this wrong? The idea that $1 of deposits create 10x as much money through FRB seems hyperbolic.
I had put this on the February links post but got there very late, and likely no one will see it, so will re-post here. I'm surprised at how absolutely no one has commented on link 18
>> "18: Thomas Piketty (plus coauthors) is the most famous historians of inequality, and says it has increased dramatically in recent decades. Now Auten & Splinter have new data(https://davidsplinter.com/AutenSplinter-Tax_Data_and_Inequality.pdf) challenging their position. Here’s Piketty etc’s response, Tyler Cowen’s commentary, and Vincent Geloso’s commentary.(https://twitter.com/VincentGeloso/status/1736483851357028857) "
(adding only two links manually because substack is impossible)
Piketty and his diatribe on inequality have been at the forefront of the left crusade on capitalism and classical liberal style economics. And no one here finds the fact that Piketty et al were substantially incorrect worth comment?
I’ve been using Yahoo Finance for a long time to keep tabs on the US stock market and get quick quotes. Even as the whole Yahoo thing was sliding into irrelevant mess of crap, with the help of adblockers the Finance page was useable.
Not anymore, with the new design it’s truly jumped the shark.
So: anybody has a recommendation for a clean-looking site for this?
Using Firefox browser on windows.
Forgive me if someone asked this question before.
For some inexplicable reason, GOOGLE is killing Google Podcasts on 2 April 2024. That's soon!
I subscribe to about a million different free podcasts. Where should I get them now? GOOGLE wants me to get them from YouTube Music but I'd rather not since they killed my prior favorite Stitcher and now are actively homiciding Google Podcasts.
Where do I go? Spotify? Apple Podcasts? Some other place?
I'm an Android / Chrome Browser person, FWIW.
Does anyone here have any memories or articles about how the Western left - the real left, I mean - reacted to the collapse of the Soviet Union? I'm finding it weirdly hard to find information about it, and I'm interested.
I like to stay clear of culture war topics, and what once was called the IWD and is now probably more clustered as the heterodox skeptics community has been too close to it for me. Coleman Hughes used to be an exception for me, but this excellent 3 part series made me update on him as well: https://radleybalko.substack.com/p/the-retconning-of-george-floyd
Is 2024 a good year to visit Israel? Due to the war, I imagine tourism is down, so there won't be crowds at the main sites, and plane tickets and hotels will be cheaper.
If any of you live in Israel or frequently visit, can you give me advice?
I write about specific high-impact ways I've improved my life. N=1, but most of them feel pretty generalizable. Every other post on philosophy or current events so you don't feel like you're being lectured.
https://youbutbetter.substack.com/
I’ve been to Manifest 2023 and can confirm it’s been great. Highly recommend everyone to get a ticket for the 2024 event.
Just up on New Savanna: Of Lit Crit “Stars” and AI “Godfathers” – In what way is Geoffrey Hinton like Jacques Derrida? https://new-savanna.blogspot.com/2024/03/of-lit-crit-stars-and-ai-godfathers-in.html
Back in 1997 David Shumway published “The Star System in Literary Studies” in PMLA. He begins with a paragraph about George Lyman Kittredge, of Harvard’s English Department at the end of the 19th and beginning of the 20th century, noting that Kittredge was unknown to the public. Here’s the first sentence of his second paragraph:
"Kittredge, who virtually founded Chaucer studies in the United States, stood at the head of a professional genealogy that controlled the field for many years after his death, but he was not a star. Nor were any of his illustrious contemporaries or near contemporaries, such as John Manly, John Livingston Lowes, and so on. Why they were not stars and Judith Butler, Jacques Derrida, Stanley Fish, Henry Louis Gates, Jr., Fredric Jameson, Gayatri Chakravorty Spivak, and other figures in the academy are is the subject of this essay."
What I’m wondering is whether or not the so-called AI “Godfathers” don’t represent a similar phenomenon in contemporary AI. Strictly speaking I believe the Godfather term applies to the three winners of the 2018 Turing Award, Yoshua Bengio, Yann Lecun, and Geoffrey Hinton, but I believe there are others in AI with a similar status, such as Ilya Sutskever, Hinton’s student and co-founder of OpenAI, Andrej Karpathy, the former director of AI at Tesla who just made waves, albeit little ones, by resigning from OpenAI, Demis Hassibis, cofounder of DeepMind, and perhaps even such figures as Nick Bostrom and Eliezer Yudkowsky, who aren’t AI researchers but are highly influential figures through their commentary. Perhaps Sam Altman, the heroic CEO who fought off a recalcitrant board, is a star as well.
Later:
Concerning the rise of literary stars, Shumway noted: “As theory has called into question the traditional means by which knowledge has been authorized, it may be that the construction of the individual personality has become an epistemological necessity.” That seems like the state of AI today. We’ve got a very complicated technology involving a blend of engineering, science, and alchemy, lacking objective knowledge. Note only that, the technology is enormously important and will change the way we live. In the absence of objective knowledge, what choice do we have but to steer by the freakin' stars?
Question for those knowledgeable about the USSR's international relations/global left:
My rough understanding is that following the dissolution of the USSR, with the exception of Cuba and North Korea, most nations/political leaders aligned with the USSR ultimately pursued their own paths. In today's context, there seems to be little that unites these countries.
However, nearly all the nations/political leaders who still broadly adhere to their far-left/USSR roots (e.g., Lula, Daniel Ortega, the ANC, etc.) continue to hold a strong animosity towards Israel, seemingly as much as they did during the USSR era. This also applies to every day Westerners who were sympathetic to the USSR.
I'm seeking a deeper understanding of this phenomenon. These nations/political leaders/people appear less focused on the USSR/far-leftism nowadays and generally exhibit significant differences in their contemporary ideologies, which makes me believe that any positions on this aren't rooted in some underlying ideology. Moreover, Israel does not seem particularly relevant to their countries to have needed to maintain a fixed position on the subject. Therefore, I'm trying to comprehend why opposition to Israel has remained so pronounced and become entrenched in this group. Is the opposition to Israel rooted in some ideology that transcends socialism...? (I'm skeptical) or is Israel merely being used as a proxy for antipathy towards the United States, which would make much more sense as something that would remain sticky? Or some third thing...
I'm curious to hear people's thoughts on this... especially if there are other similar things like opposition to Israel that the post-USSR players have remained unified on.
Scott wrote a great article on IRB's in science in 2014. Universities go way beyond what the the federal government requires, which is silly enough. And universities take 50% of grants as overhead. Are there any universities which do science like Hillsdale does humanities, intelligently with minimum administration adn overhead? Could somebody start one? I bet it would be easier to lure top scientists away with lack of bureaucracy than with high salaries.
https://www.rasmusen.org/rasmapedia/index.php?title=Best_Articles_2021#.282.29_.22My_IRB_Nightmare.2C.22_Scott_Alexander.2C_SlateStarCodex_blog_.28Sept._2014.29
Do we know when we're getting the resolutions for the 2023 predictions? Specifically for the ones who are still open on Metaculus (https://www.metaculus.com/tournament/2023-contest/). It's been already two months and if another month and a bit passes they will count for another leaderboard, which is not ideal.
In the article https://www.astralcodexten.com/p/openais-planning-for-agi-and-beyond , Scott discussed a statement by OpenAI on their future safety plans. In particular, that statement mentioned a clause from the OpenAI charter saying:
> We are concerned about late-stage AGI development becoming a competitive race without time for adequate safety precautions. Therefore, if a value-aligned, safety-conscious project comes close to building AGI before we do, we commit to stop competing with and start assisting this project. We will work out specifics in case-by-case agreements, but a typical triggering condition might be “a better-than-even chance of success in the next two years.”
Scott then asked to set some reminders to check back in March 2024 whether these plans for specific triggering conditions of the merge-and-assist clause went anywhere.
As far as I can tell after some internet search, there was no public announcement of such firmer commitments by OpenAI. There have been several further statements on their safety plans, including the Superalignment Team, Frontier Model Forum and Voluntary AI commitments (July 2023), but no improvements on the merge-and-assist mechanism (see their blog
https://openai.com/blog?topics=safety-alignment
for more safety statements and projects).
In fact, in December 2023 the company VERSES (whose Chief Science Officer is Karl Friston) published an open letter in which they announced having a pathway to AGI and asking OpenAI to trigger the above clause in their charter:
https://www.verses.ai/open-letter
Some discussion on the VERSES letter in the OpenAI Developers forum:
https://community.openai.com/t/verses-declares-path-to-agi-now-what/572636
As best as I can tell, there has been no reply from OpenAI to VERSES.
I am curious if other people also did some research and maybe found some more statements from OpenAI on the assist clause.
As I prepare to not vote in the next Irish referendum on something stupid in the constitution being replaced by something even stupider, I am confirmed in my bias - unique to myself in this country - that constitutions are largely useless pieces of paper and you can organise a society perfectly well without one.
For all its faults, and despite being the home of my eternal racial enemies, I have to admit that the U.K. gets on perfectly fine without a formal constitution and remains a relatively free and relatively prosperous country.
Meanwhile North Korea has and the Soviet Union had constitutions, the Soviet one guaranteeing absolute freedom of speech, decide amongst yourself if that worked or not.
Constitutions are either the dictatorship of the past upon the present subject to the whims of unelected justices - often leading to an effective krytocracy - or worse, the legislature uses referenda to not govern, or legislate, and which without getting into the specifics in Ireland that’s the case with the specific amendment here.
I feel the same about initiatives in US States as well. I’ve elected you to govern - don’t ask me to legislate.
So is it time for one last referendum or constitutional convention or whatever is needed in your country - to get rid of the constitution?
There are maybe three or four philosophers per century who make a difference in the world. Meanwhile, there are thousands and thousands of philosophers churning out philosophy papers.
What are those papers for? Who reads them? What are philosophers for?
https://raggedclown.substack.com/p/what-are-philosophers-for
Current implied probabilities for US presidential election, Betfair/Polymarket/Metaculus/Manifold (percentages, nearest whole number):
Trump: 47/54/50/48
Biden: 28/32/45/47
Obama: 8/5/1/1
Newsom: 4/3/1/0
Harris: 3/2/2/1
Haley: 3/1/1/0
Kennedy: 3/3/1/1
Since I last posted about this, the gap between Betfair and Polymarket on Trump has narrowed slightly, which is good to see, but the difference is still surprisingly large. One person now holds 28% of the Trump "Yes" shares on Polymarket, which means they have now bet over $1m. Biden continues to be much lower in real money markets than other prediction sites; one side must be wrong.
Betfair continues (IMO) to overvalue the outsider candidates, which probably represents small-odds bias, i.e. the difficulty of taking large "no" positions at long odds on an event 8-10 months in the future. This appears true to a lesser extent for Polymarket, which is probably because their market claims "supports negative risk (convert No shares to Yes of the other options)". For example, only 311 "no" shares seem to exist for Harris, despite there being 2,968,743 "yes" shares. Assuming this has actually been implemented correctly, it's a clever idea. It's worth noting that this means the bet totals are inaccurate: IIUC much less than $3m has been bet on Harris for example.
Weird question: a while ago, I went to see a therapist because my executive function was/is terrible and I suspected ADHD. After the therapist had dug into it, it appeared that actually I don’t have ADHD, just ASD, which also involve some executive dysfunction (which is probably made worse by eg. my low self-confidence, etc., I guess). So, I have ASD, I’m fine with that, but now my therapist says "maybe you should consider getting ADHD stimulants, those might help with your executive function".
She’s probably right, at least it makes some kind of sense, but… I can’t just go to a psychiatrist, diagnosis in hand, saying "here it says I don’t have ADHD, now gimme the ADHD drugs anyway, pretty please?", and expect it to work! Or can I? Even assuming that I can: I’ve never seen a comparison of the various kinds of stimulants there are, from Adderall to weird nootropics to very very strong coffee. Anyone has good sources on what works best / what has the best "effects on executive function : difficulty of getting it and magnitude of side effects" ratio?
Is the age of formal classroom schooling entering a long secular decline? And if so - apart from the growth of homeschooling - what could possibly replace it as the primary means by which society teaches its children? https://grahamcunningham.substack.com/p/teach-your-children-well
There isn’t much literature on this, so I’m asking here: if you were on 150mg Wellbutrin and 60 mg buspirone, would you avoid nonalcoholic beer (defined as less than 0.5% ABV)?
All literature agrees you should avoid alcohol on these medications. Studies haven’t really been done on NA beers like Athletic Brewery which are increasingly popular. Any thoughts, inferences, or guidance is appreciated.
I used to read the Economist, but lately I have less time and can only get to "Politics this week". I like it a lot, but on its own it's not worth the subscription. Does anybody know of a good replacement? Something to tell me the important news from the recent week around the world, in very brief form?
Proposition 1 adds additional criminal penalties to homelessness and nothing justifies that. The state has not shown a willingness to abide by restrictions it places on itself, so a promise to provide services in exchange for making certain things double illegal is really just a promise to make things double illegal.
Can someone who knows the science please explain to me: Does IQ matter?
Many years ago I read Stephen Gouldings The Mismeasure of Man, where he says there have been many attempts to measure human intelligence on a straight line, and they have all been flawed. And that IQ is just one of the latest of these. He said IQ had been created to separate children needing special attention from other kids, and was not meant to be used for individuals.
But in recent years, there have been many people claiming IQ is important. The chief proponent I remember is Jordan Peterson who claims IQ is the main indicator of how much you will earn, and if your IQ is less than 80 even the US Army won't hire you. This 2nd claim got some criticism online and some people said it was a made up fact.
But I dont know enough about this field--how relevant is IQ? Is it really as important as many people are claiming it to be?
A friend who is a little further left than I am but who I trust a lot has shared her picks for California ballot:
U.S. Senate: Katie Porter
Representative, CA district 12: Lateefah Simon
State Senate District 7: Jesse Arreguin
State Assembly District 14: Buffy Wicks
Alameda County Board of Supervisors District 5: John Bauters
Superior Court of Alameda County Office 12: Michael Johnson
Proposition 1: No
Alameda County Measure A: Yes
Alameda County Measure B: Yes
Oakland Measure D: Yes
I might just copy her answers unless someone here makes a good argument against.
Question for libertarian and libertarian-adjacent readers: What laws or parts of government would you want to take down first?