HN user
I think it's totally fair to be skeptical, but it's also not rare to have interventions that are astoundingly effective.
Antibiotics and vaccines may not be completely free lunches, but they're very good at what they do.
The transformer paper was published in 2017, and within 8 years (less so, if i'm being honest), we have bots that passed the Turing test. To people with shorter term memories, passing the turing test was a big deal.
My point is that even if things are pleatuing, a lot of these advancements are done in step change fashion. All it takes is one or two good insights to make massive leaps, and just because things are plateauing now, it's a bad predictor for how things will be in the future.
So healthier customers means their profit is 5% of SMALLER_NUMBER.
I don't think this is completely true right? Rather, it's more accurate to say that customers that are seen as healthier get to pay less premiums, but customers that are seen as unhealthy have to pay more.
In both scenarios, you, as the insurance company, still want to be minimizing the amount of care you actually pay for.
In other words, to maximize profits, it seems like the best customer is one that's high risk (high premiums), but less likely to require a catastrophic payout. In which case, it feels like an obese high risk patient on ozempic seems like a pretty solid deal.
can you explain this statement to me more? I think i'm missing something
hmm...doesn't this possibly incentivize ozempic subsidies even more?
If you know a "customer" of yours (an individual employee) is only going to be with you until they either change jobs or go on Medicare, then it seems the name of the game then is to make sure that nothing catastrophic happens to them until you can hand them off to someone else.
In which case, they should definitely go on ozempic. Even if the effects of ozempic immediately come off after usage, it's a short-term enough solution that benefits the insurance company, no?
this is not intuitive to me. In your description, it feels like capital more closely resembles carrots then whips?
also, the person's only source of wages isn't going to be specifically making sandwiches for just you, no?
https://www.nycp.com/elements/HA1/1345
Basically, in business and life, if you're too in love with craftsmanship of a specific function, but can't relate that well to the general goal, you're more of a liability than an asset.
hm, maybe we're misinterpreting each other's main point.
My reply was to some person who said that AI was akin to astrology, i.e. absolutely fake bullshit, which is bonkers to me.
Your reply was that AI, like 3d printing, is likely not going to be mass adopted by the average consumer, despite the hype, which i think is a reasonable prediction, and doesn't necessarily mean it won't have some valuable applications.
Maybe just agree to agree?
sorry, maybe i'm not completely understanding what you mean by "in consumer products".
reading your argument on reddit, it seems to me that you don't consider 3d printing a success because there's not one in every home...which is true.
but it feels uncreative? like, sure, just because it hasn't been mass adopted by consumers, doesn't mean there wasn't value generation done on an industrial level. you're probably using consumer products right now that have benefitted from 3d printing in some way.
ML absolutely has an applicable location, it's not nearly as vast as the hype train would say
what hype train are you referring to? i know a lot of different predictions in machine learning, so i'm curious about what you mean specifically.
It may feel like a normal conversation but is a Chinese Room on steroids.
Can you prove that humans are not chinese rooms on steroids themselves?
Sure, 3d printing has its place (mostly in making plastic parts for things) but it's hardly the revolutionary change in consumer products that it was touted to be. Instead, it's just a hobbiest toy.
how sure are you about that?
https://amfg.ai/industrial-applications-of-3d-printing-the-u...
how positive are you that some benefits in your life are not attributable to 3d-printing used behind the scenes for industrial processes?
Just like they did when ML was supposed to take over the world
how sure are you that ML is not used behind the scenes to benefit your life? do you consider features like fraud detection programs, protein-folding prediction programs to create, and spam filters valuable in and of themself?
i feel like a common consumer fallacy is that, because you don't interact with a technology in your day-to-day life, it leads you to conclude that the technology is useless.
I guarantee you that the cloud has benefitted you in some way, even though you aren't aware of the benefits of the cloud.
which hype cycles are you referring to? and, after the dust settled, do you conclusively believe nothing of value was generated from these hype cycles?
sure, but we're not discussing the outsourcing of human companionship in this context. we're discussing the capabilities of current technology.
this is a different thesis than "AI is basically bullshit astrology", so i'm not disagreeing with you.
Understanding atomic energy gave us both emission-free energy and the atomic, and you are correct that we can't necessarily where the path of AI will take us.
GPT-4o being able to describe actual events in real time is interesting, it's yet to be seen if that's useful.
sure, but my experience is that if you are able to optimize better on some previous limitation, it legitimately does open up a whole different world of usefulness.
for example, real-time processing makes me feel like universal translators are now all the more viable
i legitimately don't understand this viewpoint.
3 years ago, if you told me you could facetime with a robot, and they could describe the environment and have a "normal" conversation with me, i would be in disbelief, and assume that tech was a decade or two in the future. Even the stuff that was happening a 2 years ago felt unrealistic.
astrology is giving vague predictions like "you will be happy today". GPT-4o is describing to you actual events in real time.
This feels like an uncreative statement though, as there are many different ways to use this data.
For example, what if an engineering manager needs to convince finance for headcount to hire a devex employee?
This is a silly example, because Ilya's umbrella isn't front-end code.
To answer your question though, yes, if Ilya was Head of UX Engineering and his front-end code wasn't up to par, he should be fired.
My tongue in cheek response is that you are correct, because a century ago, WWII didn't happen yet.
Can you provably demonstrate that education quality was better a century ago? I'm not outright trying to refute you, so much as I have a belief that people tend to overstate the "good old days", so I usually prefer more concrete data points.
Have you considered that it's possible that both you and Scott are particular good at reasoning, and it's everyone else that is much worse at this stuff?
You're pointing out a bunch of errors, but it's not clear to me that they're schoolboy errors (i.e. errors a schoolboy wouldn't have made). I would also venture to guess that the average educated person would make much more egregious mistakes.
Haha yeah, I've been to rationalist meetups too, and have come to the same conclusion that I don't like them very much. Too much IQ stroking for my tastes.
Scott specifically though, has an established history of thinking through arguments that I find much better than most, and I still think someone who trains at this stuff will do better.
To extend my analogy a little further from before, if you go to a boxing gym or boxing enthusiast meetup, there will be plenty of people there who like the idea of fighting, but don't know how to fight. That doesn't mean that a trained boxer still won't be tough in the ring though.
Oh, that's a good point, and I think you're idea is a better behavior model vs. the "stick with prior" idea.
What if the conclusion forces behavior though? For voting, you can abstain from voting if you're unclear, but what if you're in a position where you have to choose two distinctly different options.
In this case, I think you pretty much have to stick with your prior, but I'm curious if you have a more nuanced approach here.
This is the correct Bayesian action: if I know that a false argument sounds just as convincing as a true argument, argument convincingness provides no evidence either way. I should ignore it and stick with my prior.
If two diametrically opposed arguments are equally convincing, it's not intuitive to me that the "correct" solution is a middle ground. Especially if both argument state that a middle ground is not possible.
One counterexample for why I think it's not always the best approach is that there are sometimes disproportionate rewards for being right when everyone else is wrong.
Doing so, however, requires you to take an idea seriously before it has gone through analyses so extensive that everyone else believes the same idea too.
If anything, Scott is underselling himself, and could probably argue circles against the average educated person.
Debates are the rationalists bread and butter, and you should expect them to be better at you the same way you would expect a trained boxer to beat you in a fistfight.
which is good, if you want to gain weight
This feels like an overly complex response to a statement that's intentionally simple though.
The whole point of Pollan's statement is to have simple guidelines that ignore optimizations in favor of directionally good flexibility, which makes me feel like Lustig either misses Pollan's point, or just believes in a completely different philosophy.
…sorry, how much exactly do you think quants make?
I think it’s a happy medium, and I don’t think the message of the article is for you. More so for the hyper ambitious people set on generating uhnw