Posts
Comments
Asteroid impacts are a prime candidate to stop global warming.
I dunno man, Randall Munroe thinks that they would cause global warming.
The nicest thing one can say about that arrangement is that it failed to start WW III
You say this like it's some kind of grudging acknowledgement, but it's actually the entire point of the structure and a Big F'n Deal. Recall that there was less than 25 years between WW1 and WW2. It's been almost 80 years without WW3, despite high tensions at various times. WW3 would have been catastrophic, and preventing it is a great accomplishment.
If that's what Quinn (comment OP) is saying then I think it's obviously wrong - people really do value the goods and services they access via the internet very highly. This leads me to believe that this is not what Quinn is saying.
What I (post author) am saying is people don't apply even a tiny fraction of the vibes that come with that high value to their actual ISP (or, analogously, airline, electric company etc).
I think the lesson of social desirability bias is that valuable services having lower status than they "ought" to is the system working as intended.
Can you elaborate? I don't understand your point because it's too compressed. I feel like I need ~3 more sentences here to get it.
Yes, my claim is that "The Comcast Problem" is the reason for hatred, as opposed a feeling more like "I really value the service but am annoyed by the customer service and pricing, so overall meh."
On a -5 to +5 scale, I'm saying they are often at like -5 when they really should be somewhere between -1 and +1.
Well, one additional factor the US has is that various veto points and power centers cycle on different time scales.
There would also be an incentive to introduce lots of meaningless elections between irrelevant (to you) alternatives in order to abstain and accrue more stored votes.
Could also be described as "There would also be an incentive to allow others to make decisions on issue that matter more to them than to you, in order to be more likely to get your way on an issue you care about."
Re-phrased that way, it's not clear to me that this is a bad thing. If they don't care about those other issues either, then you won't gain any stored votes on net relative to other voters.
Absolutely! I value your voice. But, and excuse me if this is a misread, your posts in this series read to me like you are still trying to convince yourself and/or him.
It reads like you are a sort of rationalist Martin Luther criticizing the Pope. But, like, there are already a lot of metaphorically-protestant rationalists.
I think I'm trying to make a different point than footnote 20?
It seems like you are taking me to be saying something like "You shouldn't care what EY thinks about this Trans issue because "Everybody Knows" not to take his statements on this seriously" - that's how I read FN20.
Whereas I think my point is much more general and really not specific to Trans at all - like why be so deeply invested in the contents of some one guy's mind, at all? On any issue?
EY wrote some great (book-like objects). Inspiring, even. Worldview changing. But, like, whatever his opinions are today (on any issue), my view is mostly like, who cares? Either his arguments are convincing or they aren't.
By analogy, suppose (counter factually) that I think that Barack Obama was the greatest president in history (he wasn't, but he has to be alive for this analogy to work). Does that mean that I should decide what I think about today's political and policy problems based on what Obama thinks? Such that if Obama was wrong about something, I should engage in an epic quest to Get Obama's Attention and get him to admit he's wrong? I mean, that would be ridiculous, right?
Maybe I just don't get it because I'm not part of the Berkeley Community, I just read the writing. But my immediate reaction to this is like, why does Zack care so much about what Eliezer (2024) does or does not think? Or even whether, these days, he is or is not a fraud?
Like if you thought what he wrote in 2007 was great, just listen to that? Many (all?) authors who write great books have also written worse books. Maybe Zack's opinion is falling a long way from wherever it was.
But perhaps he would be happier to adopt a more ecumenical non-Berkeley-ite stance, which I think has been common all along outside The Berkeley Community, and which is something like "Eliezer wrote some great stuff that was very influential on my thinking and that I still believe was very insightful, and I really appreciate that. I enjoy reading LW more than I think I'd enjoy the marginal alternative use of reading time, but I don't go too far out of my way to pay attention to or care about what he's up to these days." - rather than assigning himself an Epic Quest to Win This Argument.
One thing to further ponder is the extent to which systematic or repeated boundary violations can effectively amount to a dissolution. Analogous examples:
- Forcing someone to submit to multiple-times-daily injections, so far all of which have been harmless saline
- Constantly stealing objects from someone's house in a way that they don't feel like they can meaningfully accumulate personal property
- Entering a country with closed borders so frequently that its ability to enforce its immigration laws is effectively gummed up
(See this list of bar associations just for Massachusetts.)
Minor point but this is often misunderstood. These bar associations are essentially networking groups for lawyers. They are not required in order to practice. What's required to practice is bar admission which is different. There's also a federal bar admission, but that's only two, not dozens.
In a game where you play a higher number of shorter games, you can ideally have a handicap that adjusts after every game. For example, in Super Smash Bros, if you turn handicap to "auto" then the stronger player starts with damage, which (in two player) goes up 10% every time they win, and down 10% every time they lose. It gets a little more complicated in 3+ player games, and I'm not sure the exact algorithm, but it works reasonably well. Maybe something to emulate in a game where handicaps can be reasonably granular?
Yeah, I mean I guess it depends on what you mean by photorealistic. That cat has three front legs.
My general principle here is a generalization of the foundations of tort law - if you do an act that causes harm, in a way that's reasonably foreseeable, you are responsible for that. I don't think there should be a special AI exception for that, and I especially don't think there should be an open source exception to that. And I think it's very common in law for legislatures or regulators to pick out a particular subset of reasonably-foreseeable harm to prohibit in advance rather than merely to punish/compensate afterwards.
I'm not sure what "human level" means in this context because it's hard to directly compare given AI's advantages in speed, replicability, and breath of background knowledge. I think it's an empirical question whether any particular AI model is reasonably foreseeable to cause harm. And I think "enable any of [the listed] harms in a way that would be significantly more difficult to cause without access to the model" is an operationalization of foreseeability that makes sense in this context.
So with all that said, should it be illegal to effectively distribute amoral very cheap employees that it's very easy to get to cause harm? Probably. If I ran an employment agency that publicly advertised "hey my employees are super smart and will do anything you tell them, even if it's immoral or if it will help you commit crimes" then yeah I think I'd rightly have law enforcement sniffing around real quick.
Is it your view that there is a substantial list of capabilities it should be legal to freely distribute an AI model with, but which would rightly be illegal to hire a person to do?
This also would outlaw open source models at a fairly weak capabilities level.
That seems good, if those open source models would be used to enable any of the [listed] harms in a way that would be significantly more difficult to cause without access to [the open source] model. All those harms are pretty dang bad! Outside the context of AI, we go to great lengths to prevent them!
It sounds like some of those examples don't meet "in a way that would be significantly more difficult to cause without access to a covered model" - already covered by the bill.
Ah, the bill answers this question!
(n) (1) “Hazardous capability” means the capability of a covered model to be used to enable any of the following harms in a way that would be significantly more difficult to cause without access to a covered model:
(A) The creation or use of a chemical, biological, radiological, or nuclear weapon in a manner that results in mass casualties.
(B) At least five hundred million dollars ($500,000,000) of damage through cyberattacks on critical infrastructure via a single incident or multiple related incidents.
(C) At least five hundred million dollars ($500,000,000) of damage by an artificial intelligence model that autonomously engages in conduct that would violate the Penal Code if undertaken by a human.
(D) Other threats to public safety and security that are of comparable severity to the harms described in paragraphs (A) to (C), inclusive.
And "critical harm" means that same list.
Sure, but at the same time it's illegal to sell bazookas specifically because there is an unreasonable risk that a user may be able to use them to commit the obvious potential harms. So this is not some general tool-agnostic principle - it's specific to the actual tool in question.
So in this metaphor one must determine, empirically, whether any given AI product is more like a chainsaw or a bazooka. Here, the bill proposes a way to make the categorization.
This bounty might be somewhat controversial here, and I have to be honest that it almost deterred me from posting this on LessWrong, offering a bounty for a "capabilities advance"
I think that this is not a "capabilities advance" in the parlance of around here - where you are looking for ways to elicit more capabilities of existing models, rather than looking for ways to create more powerful models.
I live in the Boston area too, and this doesn't exactly sell me on going to one of those meetups :/
But more importantly to the reasons people actually buy insurance, they redistribute comparatively small amounts from the (most people) lucky to pay large amounts to the (few) unlucky. Which, behind the veil of ignorance, you would want, to ensure (insure) that your life is not ruined by bad luck.
And specifically, the risk they hedge against is usually some major risk to themselves. So insurance is similar to a social safety net in some sense. If there's a (totally made up) 1/100 lifetime chance of each person being severely injured in a car crash, and such an injury would both cost me a lot of money and a lot of earning power, then of course I'd want to insure against it. Even though the insurance company takes a cut, I'd much rather lose money on this insurance contract than collect on it. And we hope that market competition prevents the insurers from taking too big of a cut, because the insurers compete on rates. Prediction markets just don't serve this function at all.
People in this thread are focusing too much, I think, on bespoke kinds of insurance (which is most kinds of insurance), and not enough on normal everyday insurance (which is most actual insurance contracts).
A search engine that gives results like Google but before everything turned to sludge.
The thought would be that it would be the same car, but with some kind of software/hardware limit that prevents it from continuing to speed up once it reached some set speed, like 85 or 90. Not to limit the power train.
Based on the title, I thought you were going to go another direction. But isn't it insane that a typical consumer car is capable of driving significantly over 100mph? In large parts of the country, there's nowhere that it's legal to drive anywhere near that fast, and an ordinary driver will never have a legitimate reason to drive that speed. I understand why this hasn't happened, but wouldn't it also be better if normal cars just weren't capable of going over, say, 90mph?
I would say that the existence of superheroes/villains, wizards, etc would be the kind of crazy things I'm talking about. I would posit that a pretty high percent of video games (aka low-fidelity simulations) have a player who can do things easily that even the most elite athletes can't approach in real life. I'm talking about having physical abilities like 100x or 1000x average, or abilities different in kind such as the ability to fly unaided, shoot lasers from their eyes, breathe water, throw fireballs, survive dozens of gunshots, etc. That would be essentially "Spider-Man" in my analogy. But you don't see that.
Untrained men's average bench press doesn't have super reliable sources but one source I saw put it at 110 lbs. I think that's a little high, so let's call it more like 75lb. That puts the world record (unaided) bench press at 10x average - not Spider-Man/Superman/Hulk/etc territory. Similarly, average running speed is (conservatively) 5mph. Top sprint speed ever recorded is 28mph - much faster but less than 6x, not The Flash territory.
In short, there are elite athletes but no superheroes or wizards.
I...am honestly not sure. Probably mix of all? But i see the "probably not a dealbreaker" category as in the nature of "we all sometimes hurt each other, this hurts a lot, but it doesn't necessarily outweigh all the good years and forgiveness is possible" - not like it doesn't matter
Speaking only for myself, yes that's basically right. Non-monogamous behavior is evidence in favor of several bad hypotheses, but only some of which would make me mad or want to break up. Split and Commit. Things that it would be evidence of:
Dealbreaker:
- She doesn't want to be committed to me any more
Mad but not necessarily a dealbreaker over a long marriage if we can work it out:
- She wants to remain committed to me but is having some problem with our sex life and is too scared/embarrassed/confused to talk to me about it
- She wants to remain committed to me but has developed serious issues estimating and/or controlling her voluntary alcohol (or other drug) use
- We had a serious miscommunication and she honestly but unreasonably thought I had told her I was OK with whatever she did
Not mad:
- She's suffering from some kind of medical condition that causes her to act uncharacteristically or be otherwise unable to control her behavior
- She was the victim of some severe psychological or chemical manipulation
- She was suffering from physical duress/threat
- We had a serious miscommunication and, on reflection, I think it's my fault - the most reasonable interpretation of my words/acts in retrospect was that I had told her I was OK with whatever she did, even though that's not what I meant
- She didn't actually cheat at all but circumstances conspired to nonetheless lead to strong appearances in favor of a cheat hypothesis
So if I were to get some strong evidence that my wife cheated, I would want to try to collect some more evidence that would differentiate which of these nine realms (or are there others?) that we are in.
Yeah, it would be alarming! It might lead you to wonder, in addition to "should we break up", additional things like "does she have an undiagnosed brain tumor or hormone disorder", "did someone drug her", etc. I think I ultimately agree with you that it would be highly uncharacteristic behavior, and in some ways the fact that it's highly uncharacteristic is the ultimate metric we are shooting for, and the actual behavior is just an imperfect proxy for that. And then you would have to figure out what the cause of the highly uncharacteristic behavior was.
I realized I might not have been clear above, by "state" I meant "one of the fifty United States", not "the set of all stored information that influences an an Agent's actions, when combined with the environment". I think that is absurd. I agree it hasn't been shown that the other meaning of "state" is an absurd definition.
If that happened once, then my desired type of partner would want to avoid that in the future, and would avoid drinking any alcohol at future parties, and would learn to cut off interaction—rudely if necessary—if someone is making serious progress at seducing her; and if that was her response, then I'd be fine staying with her. If she didn't change her behavior and acted like that event was fine and she wouldn't mind if it recurred, then I wouldn't want to be with her anymore.
I know that @ymeskhout disagreed with this, but it seems basically right to me. For a marriage of (hopefully) many decades, I don't necessarily expect perfection at all times, but I expect significant and honest commitment. So, I wouldn't consider one mistake in several decades to be a dealbreaker, if all parties agreed it was a mistake and made active attempts to do better. IDK how long @ymeskhout has been with their partner - maybe their perspective could change over time? I have been (monogamously) married for 11 years FWIW.
I did ask one pretty-rational monogamous person where she drew the line in terms of what forms of touch counted as cheating, and it was from her that I got the "If you're asking this question then we're not compatible"
I think this is evidence in favor of the hypothesis that she wanted someone who was conceptually committed to monogamy, not just committed to monogamous behavior. For such a person, that question sounds like "I want to be as non-monogamous as possible up to some arbitrary line, and then stop, so as to avoid breaking my commitment to you. Please tell me where that line is." I think if you imagine all potentially non-monogamous-ish behaviors on a one-dimensional X axis, with some kind of intimacy-weighted frequency on the Y axis, then this question implies that your frequency graph might be flat or even increasing up until the "policy line", and then down to (hopefully) zero.
I would submit that the behavior of an actually-monogamous person would look more like exponential decay as you move right on the X-axis, and that you may not want or need a "policy line" except that, because the Y-axis is intimacy-weighted, you likely reach a point where it's not possible to engage in more than zero of that behavior while continuing the exponential decay curve.
This comment has many good questions. More generally, I suspect that for any given membrane definition, it would be relatively easy to do either or both:
A - specify multiple easily-stated ways to torture or destroy the agent without piercing the membrane; and/or
B - show that the membrane definition is totally unworkable and inconsistent with other similarly-situated agents having similar membranes.
B is there because you could get around A by saying absurd things like 'well my membrane is my entire state, if nobody pierces that then I will be safe.' If you do, then people will of course need to pierce that membrane all the time, many agents' membranes will constantly be overlapping, and the 'membrane' framework just reduces to some kind of 'implied consent' framework, at which point the 'membrane' isn't doing any work.
I suspect it's not a coincidence that this post focuses on 'membranes' in the abstract rather than committing to any particular conception of what a membrane is and what it means to pierce it. I claim this is because there cannot actually exist any even reasonably precise definition of a 'membrane' that both (a) does any useful analytical work; and (b) could come anywhere close to guaranteeing safety.
I think this misses what people find so attractive about monogamous marriage. The act of constantly comparing what one has to what one could have imposes a lot of psychic costs. Choices are Bad. Comparison is the Thief of Joy.
Better to have both parties make a socially "enforceable" commitment to stop dithering and choose. Of course you are, in a sense, actively choosing to continue to be together every day (because you could still leave), and there's a way in which that's beautiful too. But the lived experience of monogamy, for me, includes being free from the burden of choice. This is what people mean by "building a life together" that you are missing.
...doesn't that mean that this bet is only favored if you think there's at least a 40% chance of this merger going through? I wouldn't take that bet.
Thanks for the research! I'm guessing that there's probably a lot of nuance here, such as if, e.g. the President falsely accuses someone, then the false accusation is independently newsworthy and that might be protective of the media outlet who repeated it while saying that it doesn't believe the President's accusation. But I've updated my view on the core question and disendorsed my initial comment.
You could be right. I don't practice in this area and thus don't claim to have greater knowledge than you on this. I still disagree, but people should understand this is a sorta equal epistemic status disagreement.
Whether he preceded it with “Alice says” makes little difference in terms of either moral or legal responsibility.
Morally, I agree with you. Legally, I think you are not correct at least as pertains to US law, which has much higher standards to meet for defamation claims than most European countries. In the US, the truth of the statement is generally an absolute defense to liability. If I publish a story of the form "A says B committed a crime; B denies/disputes it", then in general I would not have liability if A in fact said that, because my statement was true (though A might have liability, of course).
This is part of what I'm getting at, but you seem very focused on the outside perceptions of the married people in that situation. I'm saying that the subjective experience of being one of the married people in that situation is different and, imo, usually better and more comfortable. Including in situations like a party where single people might be open to sexual attention.
I feel like, for certain coed social spaces, the cultural expectation of universal monogamy actually does (did?) a lot of work. If I (married man) am hanging out with some other married woman, we have Common Knowledge of each other's unavailability. In my subjective experience, it breaks the attraction->desire link. And it's desire that seems to add all the social tension, not attraction per se.
separate from enshittification due to capitalistic forces (changes made to attempt to please investors, create endless growth, make more money generally), are changes to apps and websites worse
I think you can't separate these things. This is a large driver of these kinds of changes, but there's no public information to definitively know which is which. So you see like 10 updates and maybe 8 of them are driven by this but you don't know which 8. And maybe 7 of the 10 updates make the product noticably worse but it's hard to know what the overlap is.
But perhaps someone who writes blog posts encouraging other people not to pay their taxes as a form of effective altruism is not your typical refuser...
They are equivalent in their effects on governmental tax receipts. Other effects vary significantly.
Also, it's not clear that governmental tax receipts actually constrain government spending, at least not directly. And if you are talking about whether your specific dollar goes to some bad effect, I think the "it's not new" flavors of MMTers* have it right - government spending creates dollars and taxation destroys dollars. So there's no more direct connection between your personal tax dollars and any particular government spending. Government creation of dollars by spending overall can be more or less inflationary depending on how many dollars government destroys by taxing.
* MMT has often been characterized as either not new or not true, and there are different claims it makes. I would say that the alleged "not new" claims are in fact true, and make no claim regarding the agreed-new but allegedly not true claims.
"You could save up to 15% or more on car insurance"?
(maybe too political but TBH the best example) "Iraq’s government openly praised the attacks of September the 11. And al Qaeda terrorists escaped from Afghanistan are known to be in Iraq."
You can’t single-handedly recreate the 1960s
One friend said that after reading about historical rates of parental supervision, she’d take her preschooler to the park and say, “Have fun, I’ll be over here reading my book.” I also try to channel the older laid-back approach to supervision of outdoor play.
But you can’t create the social environment that existed when all the kids had less supervision. This isn’t just the “someone will call the police” fear; it’s more prosaic too. At some point other parents will view you as suspect and won’t let their kids play with yours, which defeats some of the purpose.
The main thing, for me, is not that other parents will view us as suspect, but just that other kids are less likely to be there. My kids don't want to just go out and run around because the other kids aren't out there to run around with.
I don't see much similarity in the considerations that go into the question of whether to pay what one legally owes, and whether to rearrange one's affairs in a manner that results in less legal tax liability. So presenting them as two species of fundamentally the same thing doesn't help anyone understand or decide anything relevant.
On the actual moral question, I think "hey why don't we just not pay tax" is the same species of terminal misaligned EA-brain that led to SBF.
This post obfuscates more than it illuminates. People should pay all the tax they legally owe, and should be free to reconfigure their affairs to reduce the amount they legally owe, if that overall seems good/worth it to them.
Conflation of "reduce what you legally owe" and "don't pay what you legally owe" is bad and unhelpful to anyone.
We live in a society here man, c'mon.
I especially like "or could be easily modified to exhibit" here.
Generally for a court case, a generalized grievance is not enough - you need to have a concrete and particularized harm. What you are talking about sounds most analogous to "Public Nuisance" - I don't find the Wikipedia summary to be very good, and don't have time at this moment to find a better one. But the short version is, to bring a claim for Public Nuisance, you have to either be the government, or else be an individual who has an additional special harm above and beyond the general harm to the public.