Posts
Comments
Just curious, but if you found a big group house you liked where everyone had kids, would you be interested? I guess it would have to be a pretty big house.
You should probably take reverse-causation into account here. I doubt the effect of the school is nearly as strong as you think, since people who want finance jobs are drawn to the schools known for getting people finance jobs. Add to that that the schools known for certain things are the outliers. If you go to a random state school, the students are going to have much more varying interests.
Any chance you can link to that discussion? I'm really curious.
When people talk about p(doom) they generally mean the extinction risk directly from AI going rogue. The way I see it, that extinction-level risk is mostly self-replicating AI, and an AI that can design and build silicon chips (or whatever equivalent) can also build guns, and an AI designed to operate a gun doesn't seem more likely to be good at building silicon chips.
I do worry that AI in direct control of nuclear weapons would be an extinction risk, but for standard software engineering reasons (all software is terrible), not for AI-safety reasons. The good news is that I don't really think there's any good reason to put nuclear weapons directly in the hands of AI. The practical nuclear deterrent is submarines and they don't need particularly fast reactions to be effective.
While military robots might be bad for other reasons, I don't really see the path from this to doom. If AI powered weaponry doesn't work as expected, it might kill some people, but it can't repair or replicate itself or make long-term plans, so it's not really an extinction risk.
I don't think there's anything misleading about that. Building AI that kills everyone means you never get to build the immortality-granting AI.
You could imagine a similar situation in medicine: I think if we could engineer a virus that spreads rapidly among humans and rewrites our DNA to solve all of our health issues and make us smarter would be really good, and I might think it's the most important thing for the world to be working on; but at the same time, I think the number of engineered super-pandemics should remain at zero until we're very, very confident.
It's worth noticing that MIRI has been working on AI safety research (trying to speed up safe AI) for decades and only recently got into politics.
You could argue that Eliezer and some other rationalist are slowing down AGI and that's bad because they're wrong about the risks, but that's not a particularly controversial argument here (for example, see this recent highly-upvoted post). There's less (recent) posts about how great safe AGI would be, but I assume that's because it's really obvious.
I would be more worried about getting kicked out of parties because you think "the NRC is a good thing".
More seriously, your opinion on this doesn't sound very e/acc to me. Isn't their position that we should accelerate AGI even if we know it will kill everyone, because boo government yay entropy? I think rationalists generally agree that speeding up the development of AGI (that doesn't kill all of us) is extremely important, and I think a lot of us don't think current AI is particularly dangerous.
To be fair, the one-in-a-million legislators who make it to the federal level probably are very good at politics. It's kind of unreasonable to hold them the the standard of knowing (and demonstrating their knowledge of) things about economics or healthcare when their job is to win popularity contests by saying transparently ridiculous things.
I'm not downvoting because this was downvoted far enough, but downvoting doesn't mean you think the post has committed a logical fallacy. It means you want to see less of that on LessWrong. In this case, I would downvote because complaining about the voting system isn't interesting or novel.
I realized after asking that my default prompt makes ChatGPT really verbose so I changed the prompt to:
Identify types of human cells using the following marker genes. Identify one cell type for each row. Only provide the cell type name and no other commentary.
And it gave me:
- Embryonic stem cells
- Induced pluripotent stem cells
- Endoderm
- Granulosa cells
- Oocytes
- Pituitary gland cells
- Germ cells
- Leydig cells
- Neurons
- Meiotic cells
- Sertoli cells
- Neural progenitor cells
For 9 it's actually interesting that if I let it give commentary it says:
CASC3, PGAP1, SLC6A16, CNTNAP4, NPHP1 - This set of genes does not point to a well-defined cell type but could suggest Neuronal Cells or specific types of Neural Precursors based on the presence of neural development and function genes.
For what it's worth, Comcast is really, really good at providing reliable internet access (providing relatively good managed WiFi routers since WiFi is usually the worst part of the network, proactive detection of downtime and service degredation, improving latency even though it's not a 'headline number', maintaining enough slack that they hit the "up to" advertised speed close to 100% of the time, etc.). The only service issue they have is not caring up upload speeds, but there's a fundamental tradeoff with the legacy cable network and they're probably right that most people would rather have faster downloads than faster uploads (still makes me sad though).
I'm probably biased because I worked for the cable industry (around a decade ago), but purely looking at service quality, Comcast is actually very impressive.
So Comcast is stuck with zero credit for when it provides me with near-instant access to an almost infinite amount of great content (much of it for free[1]), but major blame for the small % of the time when it doesn't.
My disagreement is that I don't think people are generally upset with Comcast about internet service problems, they're upset about completely different parts of the business (billing, customer service).
I think this is fair, since "hating" a company typically has to do with how you feel about your interactions with them (do they treat you fairly, nicely, etc.), not how good they are at their jobs.
Taking this the other direction, some local ISP's provide service that isn't very "good" (using wireless tech, which has fundamental limitations, having fewer people on-call to fix problems, having fewer people to spread up-front costs to), but are very wholesome and nice to work with. Even if I choose not to use their service because of the limitations, I don't hate them because they're doing their best.
I think people hate Comcast because of their customer service and pricing, not the quality of their product. I know plenty of people who used to[1] use Comcast despite hating it because the service was so much better than the competitors.
My hometown has really good sort-of-city-provided fiber now so no one who cares uses Comcast anymore. ↩︎
One issue is figuring out who will watch the supervillain light. If we need someone monitoring everything the AI does, that puts some serious limits on what we can do with it (we can't use the AI for anything that we want to be cheaper than a human, or anything that requires superhuman response speed).
But it also creates an incentive to bring lots of annoying stuff to vote to force your political enemies to vote for it. For example, if you put "Deport all Rationalists" up for vote as often as possible, you can prevent Rationalists from voting for anything else.
If you’ve got 100-300 kilovolts coming out of a utility and it’s got to step down all the way to six volts, that’s a lot of stepping down.
We just need Nvidia to come out with chips that run on 300 kV directly.
There's an idea in security where you should avoid weak security because it lets you trick yourself into thinking you're doing something. For example, if you're not going to protect passwords, in some sense it's better to leave them completely plaintext instead of hashing them with MD5. At least in the plaintext case you know you're not protecting them (and won't accidentally do something unsafe with it on the assumption that it's already protected by being hashed).
I feel like this is a case like that:
- If you don't care if these become public, consider just making it public.
- If you don't think they should be public, use something that guarantees that they're not (like the random ID solution)
The solution you proposed is better than nothing and might protect some email addresses in some cases, but it begs the questions: If you need to protect these sometimes, why not all the time; and if not protecting them sometimes is ok, why bother at all?
(I should say though that there are benefits to making data annoying to access, like that your scheme will protect the data from casual snoopers, and prevent it from being crawled by search engines unless someone goes to the trouble of de-anonymizing and reposting it. My point is mostly just that you should ask if you're ok with it becoming entirely public or not)
Would plans to stop a rogue AI from hacking things be any different than the normal work people do to prevent hacking?
I realized after writing this that you meant that people's email addresses are private but their scores are public if you know their email. I'd default to not exposing people's participation and scores unless they expected that to happen, but maybe that's less of an issue than I was thinking. The predictability of LessWrong emails still would expose a lot of email addresses.
I'd still recommend the random ID solution though since it's trivial to reason about (it's basically a one-time-pad).
I think this would provide security against people casually accessing each other's scores but wouldn't provide much protection against a determined attacker.
Some problems:
- There's no protection at all for someone's scores if the attacker knows their email address (and email addresses aren't secret)
- It's probably not that hard to build or acquire a list of LessWrong users' email addresses
- Even if you just brute-force this, there are probably patterns in LessWrong users' email addresses that make them distinguishable from random email addresses (more likely to be @somerationalistgroup.com, @gmail, recognizably American, nerdy, etc.).
A better solution:
- Generate a random ID for each user and add it to your data
- Email users their random ID
- Publish the data with emails removed
(And remove anything else that could be used to reconstruct users, like jobs/locations/etc. if relevant)
That actually would explain a lot about art values.
Jane Street is a pretty extreme comparison. An easier one is that a good software engineer at Google can, in their late 20's, make 2x what a tenured professor makes by the end of their career, with similar or better work/life balance. Tenure becomes irrelevant when you can retire by 40.
I suspect the discrepencies in our land value vs improvement value numbers have to do with where the land is and how efficiently it's used. If you have a single family home in San Francisco, most of the value will be land, but it seems undesirable that your proposed tax would very heavily penalize anyone who tries to turn a single-family house in SF into a skyscraper (with a much lower land/improvement ratio).
As for skyscrapers, the interesting thing about this proposal is that hard-to-remove amendments essentially become land. For example, if you made a plot of land fertile, that improvement is difficult/undesirable to remove, so when you go to sell it, the owner pays for it as if it were land. I'll tackle this more in the second post.
Taxing improvements (discouraging people from improving land) seems like the exactly opposite of what a land value tax is supposed to do. I look forward to how you address this in the second post thogh.
I actually thought Richard's post was a joke until I read this. I'm impressed.
Since bidders know they can get away with buying the home cheaper than the true value, they will increase their bids. Without taxes, the bidders would increase their bids by almost the V/2 profit they can expect to make. However, they won’t increase their bids fully, since they will be taxed based on the higher bids. Notice that from the sellers perspective, they get roughly the true value of the house, it is just that part of the home value comes from the land bid while another part comes from the home sale. From the seller’s perspective, this looks like a small “tax” on the value of their home. From the buyers perspective, they will be taxed on the portion of the home price that they included in their land bid.
Isn't this kind of fatal to the whole idea?
The reason to use auctions is to get a more accurate price, but the "land" value you're assessing is expected to be off by half the value of all improvements. Given that [it costs around $400k to build a 2,500 sqft house and $100k to buy the land](https://www.ramseysolutions.com/real-estate/how-much-does-it-cost-to-build-a-house) (which incidentally is fairly close to my insurance company's rebuild cost and the assessed land value of my house), the $300k you'd assess land taxes against is 2/3 improvement. It would get worse with something like a skyscraper where the building is orders of magnitude more expensive than the land it sits on.
I can't test this right now, but I wonder if part of the problem is that you're prompting it to have fun and (implicitly) to tell you a story, not to do logical thinking. I wonder if a "I was reading about this historical thing..." prompt, or a fully-modern prompt would help, but if you make it too realistic ChatGPT's "don't tell users to kill people" behavior will take over.
So on atheism, it’s really hard to see how Beth 2 people could possibly exist. But if fewer than Beth 2 people exist, then 0% of possible people exist, which would make the odds of my existence in particular zero. I’m not special—if 0% of possible people exist, it’s ridiculously unlikely I’d be one of the lucky few that exist.
This has been discussed on LessWrong several times recently, but you're you're using the wrong denominator when deciding that your existence in particular is unlikely compared to all other humans.
On way to demonstrate this: If I flip 256 coins, the probability that I get HTTTTTTTHTTHHTTTHTTTHTHTHTTHHTTTTTTHHHTHTTHTTTHHTTTTHTHTHHHHHTTTTHTHHHHTHHHHHHHTTTTHHTHHHTHTTTTTHTTTHTTHHHTHHHTHHTHTHTHTHTHHTHTHTTHTHHTTHTHTTHHHHHTTTTTTHHTHTTTTTHHTHHTTHTTHHTTTHTTHTHTTHHHTTHHHTHTTHHTTHTTTHTHHHTHHTHHHHTHHTHHHTHHHHTTHTTHTHHTHTTHTHHTTHHTTHHTH is so small that I'm (much) more likely to randomly select a particular particle in the universe than to happen to get that sequence of heads/tails, but I'm not an all-powerful God who flips unbelievably many coins; I only ran my coin-flipping script once (ok, like 3 times because my script had subtle errors).
Other recent posts have covered this better and in more detail, but I can't find them at the moment.
Why think that my existence is very likely if there’s a God? Simple: God would create all possible people. It’s good to create a person and give them a good life. There’s nothing stopping God from creating any person, so he’d make them all. God would make anything that’s worth making, and every person is worth making, so God would make every person.
It seems like the world we inhabit argues against this. I'd expect there to be a lot more people (who are on-average happier) if a pro-natalist benevolent-and-aware-of-humans God existed.
You could argue that God did create the pro-natalist utopia, then the slightly-less pro-natalist utopia, and so-on down to our weirdly-empty and unhappy world, but then your anthropic argument breaks down (if you randomly select a human, ~0% of them are not in the pro-natalist utopia(s)).
If the expected number of future paperclips is astronomical, shouldn't you be short paperclip futures, not long?
... I'll see myself out.
I just got around to reading this in my RSS reader (I rarely have time to read something of this length), and I was sad to see that the original was removed. I don't really have anything important to say but I think these posts have been interesting.
It seems like a major problem that you don't consider pleasure/suffering at all. Wouldn't this framework consider it extremely good if you could increase everyone's lifespan by one day in exchange for making them feel intense pain at every moment until then?
I don't know the right keywords to search for it, but there's a related network security technique where you monitor your incoming/outgoing bandwidth ratio and shut things down if it suddenly changes. I think it's typically done by heuristics though.
Yeah, it turns out to be really important that psychopaths usually aren't very smart. Making it easier for dumb people to do really bad things seems.. bad.
I'm not sure if all of these descriptions would pass the ideological turning test, but I upvoted because I think the post is broadly right and interesting.
(For example, Object-Level theory is the closest to how I feel about this, but I don't think thinking about rationality is a failure mode exactly (although there is a type of thinking about rationality that I do think is a failure mode).)
I personally think that the chance that covid-19 was created in a lab in Wuhan is exceptionally high, perhaps 93%, and there are various skeptical experts who think it is now beyond reasonable doubt that the Wuhan Lab created covid-19.
This might have already been covered somewhere, but I'm curious what makes you think COVID-19 was created in a lab and not a natural virus leaked while they were studying it.
Update: Roko wrote a whole post about this.
I'll make a note to blur them or something, but I suspect anyone with the motivation to copy my keys from the internet could probably pick any of these locks too.
Oh yeah, that's a good idea. You'd need to find a sufficiently small carabiner (most actually-good ones are pretty big), and I think you'd need to put the keys on larger rings than I used to be able to get a carabiner through them. I think if you wanted a stronger system that would work, although it might end up being bulkier.
I'm not really worried about strength myself though. The carabiners are probably not as strong as the listing says (15 kg max weight), but I only need them to hold the weight of a couple keys.
If your point is just that "some expensive things aren't just status symbols, and worth the price even to the merely middle class," fine, I agree with that in general, though perhaps not in the specifics.
This is how I read the argument: Hiring a house cleaner is actually a reasonable thing for a middle-class American to do. Note that "middle class American" is still objectively ridiculously rich.
I do think there's something weird about treating $25k+ cars and hundreds of dollars per month on restaurants and alcohol normal, but drawing the line at $100 per month for house cleaning.
It seems worth calling out that Scott isn't saying that orthagonality is impossible, just claiming that it's harder than non-orthagonality:
Yes, there could be a superintelligence that cared for nothing but maximizing paperclips—in the same way that there exist humans with 180 IQs, who’ve mastered philosophy and literature and science as well as any of us, but who now mostly care about maximizing their orgasms or their heroin intake. But, like, that’s a nontrivial achievement! When intelligence and goals are that orthogonal, there was normally some effort spent prying them apart.
So I think the claim regarding your graphs is that it's easier to build above the line than below, not that that it's impossible.
Why do you think the link to investment strategy is silly? I just found this article last night and thought it was really useful, so if it's giving me bad intuitions that would be good to know.
Isn't real life somewhat like games #3 and #5 (the game doesn't go on forever, but the "dealer" decides when you die quit), and applying the trick from game #6 really does help?
Sorry, I wasn't sure the right way to do a link post without just copying the whole thing (which I don't have permission to do) but there's way more in the link. I tried to edit it to make it more obvious that this is an excerpt.
The game you want to play is #5 I think:
Start with $100 and play with the same rules as game #3, but this time, you get to decide what percent of your payout is wagered. I still decide when to quit.
Edit: I decided to remove the whole excerpt.
I don't think I ever ran into that when I was younger. Meeting in houses is the original way Christians met, so I think it would be weird to complain about it. I found it pretty common for people to make fun of the opposite. If you're spending your church money on a big fancy building, does that really show your dedication to church teachings like charity*?
Also, people might accuse a really small church group of being culty, but a small church group with a big fancy building feels much cultier than the same group meeting in a house.
I was only really exposed to Evangelical Christianity so it's possible this is very different among other groups like Catholics.
* Churches typically justify this in terms of practicality (more spaces to work with) and marketing evangelism.
I went to a bunch of churches when I was younger, and I think the types of locations you listed are actually in a fairly typical order for starting a new church:
- The founding members meet in someone's house. One nice thing regarding space is that the more of your founders have kids, the more likely it is that one or more has a house big enough to host this. I think at this stage it's common to rotate though a small set of locations though.
- Once you're too big to meet at a house, rent a cheap event space. The church I went to for most of my life rented the same room in a senior citizen center every Sunday, except around once a year we couldn't get it and would go to the park.
- At some point if you have enough money, you buy a whole building because this gives you more options (ability to customize and use it whenever you want).
For renting, I wonder if part of it is just calling places up to see what combo deals they can give you. When renting the space at the senior citizen center, my church had access to one big room with a stage, plus two small classrooms. I'm not sure if we got lucky and this was the only place like that, or if it's actually a reasonably common set of needs.
This also makes me wonder if combination daycare/coworking spaces exist (and if they don't can someone get me Masayoshi Son's phone number because I have a tech company idea?).
For the record, I'm not arguing against a land value tax in general. I actually think an LVT is reasonable idea if you can actually figure out how to determine land value. I just think this particular argument for an LVT has in incorrect premise, and the links used to support it don't actually support it.
I think the two cruxes of our disagreement are, first that I think you are saying that you know why homes are vacant. These two quotes claim that houses are being left vacant for speculation:
At the same time, vast brownfield sites which could be developed are left unused and many of the properties that are built are simply left empty, and held as a speculative investment.
And here:
Landlords may be unproductive, but at least they allow people to live and work on the land they own. Some land owners do not even rent out their land: they simply sit on it and wait for the price to increase. This behaviour is known as speculation and as a result, there are almost 700,000 empty homes throughout the UK, and over 150 million square feet of unused retail space.
Second, I think the reason matters. If the houses are being held vacant for a reason other than speculation, in undermines the argument for why an LVT would change that. An LVT would affect the vacancy rate if (1) it incentivizes people to rent out more houses and (2) it's possible to rent out more houses.
I think (1) is not a meaningful difference given that investors are already strongly incentivized to rent out properties they own (they get rent!), and I think (2) is mostly wrong and most of the empty houses are empty because remodeling, sales, and finding renters takes time (around 2/3rds of these houses are empty for less than 6 months) or because no one wants them (the houses are in the wrong place or not liveable).
This article summarizes the important data points here. Among other things, note that long-term vacancy rate in London is 0.4%, compared to 2.7% for England as a whole, and that the number of vacant properties has been doing down over time.
Regarding your comments, I don't think vacation property owners are relevant here since I'm disputing your claim around land held for speculation, and vacation properties are held for their owners' use. This effects the incentives since vacation property owners are already forgoing rent which is worth more than any plausible LVT (although an LVT might push some of them over the edge to rent/sell).
In short, I don't dispute the existence of empty homes; I dispute that they're empty for the reasons you claim, and as a result, I don't think an LVT will meaningfully convert those into useable houses.
I expect that your next response will be something like "yes, but a not-literally-zero amount of houses will convert from vacant to not-vacant with an LVT", and I agree that the number will probably not be literally-zero, but I think that number will not be meaningful, and that this isn't a free win with no downsides.
Among other reasons, it's important to realize that the real business a lot of house builders are in is land speculation (even better than waiting for someone else to develop your empty land is just doing it yourself), so it's unclear if destroying that business would help or hurt housing supply. There's also the political angle, that "housing shortages are caused by landlords" is used by NIMBYs to redirect away from real solutions (building more housing), so repeating their untrue claims in service of a minor win is counterproductive.
Ah I missed that. Thanks!
I still think it's a problem that this argument rests on the idea that investors are irrationally not renting land they own, but you don't provide any evidence for that.
As I try to look into this more, I'm also finding that the vacancy rate seems really low in England. 676,304 vacant homes / 24.9 million total homes gives a vacancy rate of 2.7%, which is lower than every city in the United States except for Gilbert, AZ according to this article, although it's possible vacancy is being defined differently?
You claim in two places that empty land is being held for speculative purposes. This doesn't make any sense to me (why would I refuse to rent my properties to push up the value of someone else's properties?), so I followed your links to investigate. The first link seems to just be a general purpose NIMBY article arguing against redevelopement, and I didn't see anything about keeping houses vacant. The second link is to a Scottish political campaign that doesn't claim to know why the houses are empty (at least on this page) and doesn't contain the 700,000 number in the link text (the linked political campaign claims 46,000 in Scotland and doesn't seem to say anything about the UK).
I think this answer is missing the point of the question. The deficit gets paid either way, it's just a question of whether it's paid via money printing or taxes. Ignoring money for a moment, any time the government uses resources to do stuff, someone else has to not use those resources. Money is system for allocating those resources, but changes in the money supply can't (directly) change the amount of resources available, and in a frictionless, spherical economy, there's no difference between taxes and inflation.
I thought this was useful and I'm interested to see what the next parts are.
My feedback is that the post would be improved by skipping the disclaimer and introduction. The post is already long and spending 1/5th of it on meta commentary will likely cause some people to bounce off before they get to the content.
It also feels kind of like a bait and switch because the post isn't about photographic memories, but maybe you're building up to that?
You also don't need to worry about being new to the site in general. The kind of new-person content that tends to get downvoted is usually crank theories or people being condescending.
I haven't finished reading it, but I wonder how A City on Mars would compare. I think the authors make some of the same arguments (including that space settlement probably won't be more free than Earth), but they're arguing from the perspective of people who think space settlements would be awesome.
In previous studies, we consistently find an equalizing effect from use of LLMs. High performers improve, but low performers improve a lot more.
I was just talking to a math teacher last night about something similar. He was talking about how COVID really hurt math learning and the lowest performers aren't recovering from it (doing even worse than pre-COVID low-performers). I had been talking to him about how I use ChatGPT to learn things (and find it particularly helpful for math), so I asked if he thought this kind of thing might help narrow the gap again.
His answer is that he thinks it will increase the gap, since the kids who would actually sit down and ask an AI questions and drill down into anything they're confused about are already doing fine (better than pre-COVID since they don't have to wait for the slow kids as much), and the kids who are having trouble wouldn't use it. Also, the benefit is that AI can immediately answer questions about the part that you, personally, are confused about or interested in, so neither of us thing it would be that helpful to try to force kids to use it if they don't want to.
(Schools also have annoying but reasonable concerns that if you tell kids to use the magic machine that can either help them learn faster or just do their homework for them, many of the kids will not use the machine in the way you're hoping for)