Posts
Comments
I brought it up with him again, and my father backpedaled and said he was mostly making educated guesses on limited information, that he knows that he really doesn't know very much about current AI, and isn't interested enough to talk to strangers online - he's in his 70s and figures that if AI does eventually destroy the world it probably won't be in his own lifetime. :/
Representative democracy can only last so long as people prefer losing an election to fighting a civil war.
He might also argue "even if you can match a human brain with a billion dollar supercomputer, it still takes a billion dollar supercomputer to run your AI, and you can make, train, and hire an awful lot of humans for a billion dollars."
Because there were enough people selling for prices lower than $40 to satisfy the demand for greater fools?
Also, stocks can be sold short if the price goes too high.
Yes, I know.
My father thinks that ASI is going to be impractical to achieve with silicon CMOS chips because Moore's law is eventually going to hit fundamental limits - such as the thickness of individual atoms - and the hardware required to create it would end up "requiring a supercomputer the size of the Empire State Building and consume as much electricity as all of New York City".
Needless to say, he has very long timelines for generally superhuman AGI. He doesn't rule out that another computing technology could replace silicon CMOS, he just doesn't think it would be practical unless that happens.
My father is usually a very smart and rational person (he is a retired professor of electrical engineering) and he loves arguing, and I suspect that he is seriously overestimating the computing hardware it would take to match a human brain. Would anyone here be interested in talking to him about it? Let me know and I'll put you in touch.
Update: My father later backpedaled and said he was mostly making educated guesses on limited information, that he knows that he really doesn't know very much about current AI, and isn't interested enough to talk to strangers online - he's in his 70s and if AI does eventually destroy the world it probably won't be in his own lifetime. :/
If it's 1950, is having gay sex unwholesome?
(yes I know you've being ironic)
Well, trade does have a more zero-sum character when both sides of the trade have the same preferences, but if you can credibly claim to have different preferences, you're also in a better position to convince the person on the other side of the trade that you're not trying to offer them a bad deal. (For example, if you're selling stock because you want to spend the money, you don't care if you disagree with someone about what the stock will be worth in the future; you just want to sell it for the best offer you can get right now.)
I think you missed the point of the Laffy Taffy example. He got the flavor he didn't like because he'd been systematically eating the ones he did like while leaving the flavor he didn't like in the bowl. (Or his friend wasn't actually picking at random.)
I imagine a criminal defense attorney gets lied to more than Dr. House.
Out of context, I could totally believe someone would use that name for a chat room as a joke. Then again, I'm the kind of guy who can barely keep himself from offering to tell bomb jokes to airport security.
Ah. It turns out that I was mistaken in thinking that the 5th Amendment guaranteed the right to refuse to testify against one's spouse; the text of the amendment doesn't mention spouses at all. (Mandela effect strikes again?)
That's surprising and I think I must be missing some context. Random Googling seems to suggest that spousal testimonial privilege applies to events before the marriage, at least in federal court - if he did legally marry his girlfriend, she shouldn't have had to testify at his trial if she didn't want to. Different states do treat spousal privilege differently, though, but am I missing something else? Did the police learn something from the girlfriend before the marriage that the prosecution can use in court without having her testify?
What it tends to boil down to is that they don't trust me to be their criminal co-conspirator
Yeah, as a certain TV character said, they don't want a criminal lawyer, they want a criminal lawyer.
I've heard the opposite - a dead shooting victim means there aren't any witnesses to contradict your story.
Indeed, students in physics lab courses violate the laws of physics all the time. There must be a way we can exploit this, just like we do the buttered cat paradox.
I do not necessarily agree with every claim made here, but I do agree that it would be best if the United States and Europe supply Ukraine with weapons and ammunition to the greatest extent possible. Besides voting Democrat in the next election, is there anything else that a random American with no significant political connections can do that would be useful?
People are frequently idiots. They can also be idiots who also don't trust their public defender - or know how to trust a public defender.
The person asking the question has not told us if the AIs have predicted the launch or not.
In the short run, I have my hands full with keeping up with household chores, walking my dog, and visiting and being an advocate for my severely ill wife, who has been in hospitals almost continuously since December 2022. I don't have a job (besides taking care of my family's rental property) and don't think I could handle one right now.
Most of the actual goals I manage to set for myself and achieve involve playing video games. :/
Many of the alleged "conspiracies" that literal conspiracy theories propose are highly improbable. The more people involved in a secret conspiracy, the more likely it is for someone to leak the secret, either deliberately or accidentally. In order to maintain a conspiracy, it either has to be very small ("three can keep a secret, if two are dead"), or it has to be the case that most people, if they learned the secret, would agree that it should be kept secret; it has to be something closer to a Benevolent Conspiracy instead of an evil one.
One of the most successful large-scale "conspiracies" in recent history was the Manhattan Project, and it was able to be kept secret mostly because anyone who did learn about it would, by default, want to keep the secret. If you picked Americans at random until you found someone who would be willing to betray the Manhattan Project to the Axis powers, you'd have to go through a hell of a lot of people before you found one; despite the large number of people involved, nobody told the Germans or the Japanese. On the other hand, if you picked Americans at random and asked them if it should be kept secret from our allies in the Soviet Union, the percentage of people who would disagree would be much higher - and someone actually did leak the details to the Soviet Union.
Sometimes the answer is "absolutely yes." For example, I'd love to be able to understand Japanese, but I'm not about to dedicate a year or more of my life to studying the language in order to do it. (As I've mentioned before, learning foreign language vocabulary is relatively difficult for me, because there's no way to use partial knowledge to recover something you can't quite remember; knowing that "azul" means "blue" and "rojo" means red doesn't help me remember that "verde" means green. I have to resort to brute force memorization, and I hate it.)
Another thing that I might like to do but I'm not sure is worth trying is making my own game using something like RPG Maker. I imagine that it would take a year or more to go from where I am to a working game I'd be satisfied with, and I don't know if it would "pay off" - even if I did make a really amazing game, how many people would ever play it?
As Mark Twain supposedly once said: it's not what you don't know that gets you into trouble, it's what you know that just ain't so!
This is exactly why I often don't look at object level arguments. If I don't have an actual subject matter expert on hand, an argument based on unrefuted bullshit and blatant lies will seem just as plausible as one based on the actual truth. So I just nod, say "Yes, that sounds convincing, but I don't know enough to be able to tell if there's something wrong with it, so you'll have to take it up with the mainstream if you want me to believe you.
This is often my problem - I think "I could probably do it if I really wanted to" and "I'm not doing it", and conclude "Therefore I don't really want to do it (strongly enough)."
Your frequentism is showing. Bayesian probabilities are subjective credences, not objective features of the universe.
If it's God's job, then it's definitely not going to get done, because something that doesn't exist can't do anything.
Yes, the question does not consider the possibility that Alice was playing chess against a non-human opponent or against an opponent that is not in the room.
The Chicken, aka Hawk/Dove game, is a pain in the neck to analyze from a "naive decision theory" perspective.
Level Zero analysis:
The best response to Dove is Hawk. If my opponent is going to play Dove, I want to play Hawk.
The least bad response to Hawk is Dove. If my opponent is going to play Hawk, I want to play Dove.
Going meta:
If I can predict that my opponent plays Dove in response to HawkBot, I want to self-modify into HawkBot and trap them in the Nash equilibrium that's good for me.
If my opponent is HawkBot, I want to yield and play Dove.
Going more meta:
If my opponents predict that I'll play Dove in response to HawkBot, then they'll self-modify into HawkBot and trap me in the Nash equilibrium that's bad for me. So I don't want to play Dove in response to HawkBot. But if I play Hawk in response to HawkBot, I'm basically HawkBot myself, and the HawkBot vs HawkBot matchup leads to the worst outcome possible. So it can't be right to play Dove, and it can't be right to play Hawk either.
Which leaves me with only one possible conclusion:
Well, you're filtering on both "can afford to pay for a workshop" and "wants to attend a workshop thay charges that much"...
There are many things that bother me that I can't fix because someone else prefers them the way they are.
I'm reminded of a movie quote: "Don't have a fallback plan. If you have a fallback plan, you'll fall back."
There are also stories of military leaders burning their ships after landing, making retreat impossible and leaving the only possibilities as victory or death.
I think the VCs aren't doubting your competence, but rather your commitment.
I know how I would play against myself. First, get me absorbed in a fiction so that I'm making decisions "in character" rather than as someone trying to win a game. Then appeal to my sense of morality and push at me from whatever other angles I can think of. I don't actually know if I would beat myself as the AI player, but I do think I could lose as the Gatekeeper if I didn't resort to acting out-of-character. Like, I probably could pretend to be an idiot or a crazy person and troll someone for two hours, but what would be the point?
"Sci-fi risks? You mean, like, nuclear weapons in the 1930s?"
Well yeah, but some people just put a fake label on their product and call it a day. UL can't stop outright fraud.
Does Siemens or GE or any company with assets and a reputation longer than 5 years sell such a fuse? That would be an example of a market mechanism that doesn't stop Amazon from selling it, but would help prevent savvy buyers from purchasing it.
What happened on Amazon is that items advertised as being sold by Amazon itself wouid turn out to be Chinese counterfeits. You could order a thingamajig made by Siemens and get something with the name "Siemens" on it that wasn't actually made by Siemens.
This is what I've heard that Amazon did, and I don't know whether or not they still do:
Amazon has a "Fulfilled by Amazon" service that third party sellers can use, in which they ship products to Amazon, and when someone orders the product from that seller, Amazon will then ship them the product. However, the Amazon warehouse doesn't keep track of which seller sent in which item: if there are multiple sellers that use Fulfilled by Amazon to sell a Siemens Thingamajig, they all get marked with the same ID number and put in the same bin. If Amazon itself also sells Siemens Thingamajigs directly, its own stock gets put in the bin with the ones from third party sellers. So, if a third party seller opts to use "Fulfilled by Amazon" and ships Amazon a counterfeit Siemens Thingamajig, someone ordering a Siemens Thingamajig from Amazon directly could end up with the counterfeit that the third party seller had put up for sale.
Pharmaceutical companies won't go and release hundreds of dud or dangerous drugs just because they can.
At least some of them would, because there are, in fact, companies that do this in the real world. They simply use a different brand name and call them "dietary supplements" rather than drugs.
Not to mention the actual quackery that's slipped through the cracks anyway, such as the homeopathic "treatments" that people are still somehow allowed to sell.
The New Jersey Turnpike actually does have variable speed limit signs.
What are your options if you fail the medical exam because you have cancer or something?
Well, since I like cats a lot and, when I was single, would have preferred a woman who liked cats too, maybe I could have filtered out some potentially bad matches this way? (I didn't actually have a cat, though, for various reasons.)
One thing that, in hindsight, turned out to be very bad advice for me in particular was "get your house in order before trying to date" - it wasn't until my future wife made a move on me that I was able to make my own life and mental health get better.
I heard that, on old OKCupid, when you ranked someone's appearance as 5 out of 5, it would tell that person that you did so - which would have the effect of discouraging women from giving a 5 out of 5 to someone that they didn't also want to send a giant "I THINK YOU'RE HOT" signal to.
9s and 10s may also be people who are high school or college age (or how one remembers one's high school or college peers as looking) and thus not suitable for an adult to actually date.
https://en.wikipedia.org/wiki/Sociosexuality
It's a term in psychology.
There's a theory that one thing that women find sexually attractive in men is having a high position in the (male) status hierarchy. Consider Christian Grey, powerful billionaire, and also the many older (and old-fashioned) stories in which the heroine's desirable love interest is a Prince. As Henry Kissinger supposedly said, "Power is the ultimate aphrodisiac"...
Minor exception to Taleb's Surgeon: if charisma is directly relevant to job performance (as in sales, for example), feel free to take the more charismatic candidate.
EDT chokes on Simpson's Paradox - specifically, the "Kidney Stone Treatment" example. EDT will only look at the combined data and ignore the confounding variable (the size of the kidney stones), and end up choosing the worse treatment. Which treatment you get doesn't change whether your kidney stone is large or small, but EDT will make decisions as though it does.
If anyone here has a better phrasing for something in the two paragraphs, feel free to let me know. I'm hoping for something that people can link to, copy/paste, or paraphrase out loud when someone asks why we think AI risk is a real thing.
Part of the problem is that humans themselves are often bad at knowing what they want. :/