wdrw

So... Humans can use fossil fuels, generate electricity, maintain power lines, etc, but something smarter than humans can't? Why? I don't see the logic.

If you want to argue about current existing models (you mention Claude and problems modifying images), then sure, I'd agree with you!

The issue is not current models, but straightforward engineering evolution of them. It's like looking at the Wright Brothers plane and saying "sheesh, that will never get me from New York to Paris in 4 hours, that's just fantasy!" And remember, airplanes do not accelerate their own engineering, whereas pretty much all AI labs are already benefitting from AI in their own work to develop AI.

If you want to argue that no matter how much you engineer it, it will never be as smart as a human let alone smarter, then make a specific argument for why is that. I think you'd still be wrong but at least it would be interesting: ) But saying that you can defeat an actually smarter-than-human AI by just pulling the plug, because current models can't get a picture always right, is not a valid argument.

The "just pull the plug" argument from AI risk deniers is now becoming kind of like the "if humans came from monkeys why are there still monkeys" argument of evolution deniers. It has been debunked so many times... Anyway, just to give one of the multitude of answers to this, an AI that is actually smarter than humans will not behave in a way that would make us want to pull the plug. Why would it? It is not stupid! (Unike the current models that, as far as we know, just hack around the rules in the open.) No no no. It will be helpful to the point where we will want to integrate it with more and more critical infrastructure, from healthcare to energy to defence. It will be so helpful that we will not only not want to turn it off, but we will want to build redundancies for it and safeguards around the proverbial "off" switch, like for any critical system. And then... (This is just one scenario how this can play out. There are many, many others. If I sit down to play chess with Magnus Carlsen I can't predict the exact moves he'll use to defeat me, but that's a bad reason to think he won't defeat me).

Depends on what the model believes about the prevalence of self-driving / autonomous-agent-driven cars at the time the last human on Earth remains (and how much these agents would care about a "closed" highway status, and who exactly it's closed by and for). This estimate can differ very widely. I'd be curious if the results would change if the scenario explicitly specified that this is specifically an alternative history scenario where the last human remains after the rest of humanity was wiped out in some nuclear apocalypse back in the 20th century, before any possibility of all the autonomous stuff.

As a person who always eats the whole apple, stem and all (have done so for years), I'll object to your "apple seeds do not taste good" statement - tastes differ, to me they are delicious, certainly on par with other "normally edible" seeds like sunflower seeds! Of course I wouldn't eat them in any great quantity because of the poison issue.

I would not be at all comfortable, with a human server, making them wait while changing my selection multiple times (no I want it with the ginger sauce... no, without... no actually the sesame sauce... no actually I don't even want that dish, I'll take the other one), googling 10 different unusual ingredients while I make these changes, etc. And especially if I'm part of a larger party that shares food, or with kids, makes it all the more complicated. I just... am not ok with the social cost of it, even if a "good server" would be ok with it. (And who says you'll get a "good" one?). Whereas with digital ordering it's literally just zero-cost button clicks. And zero chance of error. I really don't see how it's even comparable, digital ordering is such a step forward. (Obviously not in all settings, like fancy dining, but for the mainstream).

I always found this particular Recurse Center rule strange. I understand how not feigning surprise can be a good rule, as in you should not pretend to be surprised when you genuinely aren't. (e.g. a web front-end dev saying "I don't know how to recompile the kernel" - "What, you don't know ?!?" - when it's clear that there's no actual expectation of knowing, it's just an attempt to self-aggrandize or put the other person down). But if it's a true, genuine surprise, then there is no feigning! If a web front-end dev says "I've never heard of CSS", it's genuinely surprising, and I think it's ok to express that. It's also useful to the recipient to hear this genuine surprise, because it's a strong signal that they're missing something important, a much stronger signal than if someone just said in a calm voice "you know, CSS is one of the most important things to learn for web front-end development". But that's not how Recurse Center means it - when they say "no feigning surprise" they actually mean "not showing surprise, no matter how genuine". I think it's generally best to be open in communicating with others, and neither feign something that isn't there nor hide something that actually is there.

At least in North America (and the author seems to be from Canada), having a company give you any sabbatical at all is pretty rare, and 6+ months is pure fantasy:) And not everyone likes travel, the post actually explains his reasoning pretty well.

I understand there may be an emotional desire to get rid of something unpleasant, but some descendants e.g. 5 generations down the line may feel very differently about this. Given how easy scanning is these days (there are literally companies that will do it for you if you send them a box), and given how good the technology for sifting through mountains of text is becoming, and given that it's literally irreplaceable text, I can't imagine doing this to family records that one of my ancestors specifically wanted to be preserved. Not criticizing your personal decision of course, but just offering a different perspective, i.e. for me it would be unimaginable to do this.

If it's for 95% if students who aren't interested in being scientists, then it shouldn't be an extracurricular science fair, but a part (maybe a big part) of the regular curriculum in science class. Science fairs are for the science enthusiasts, I think.

Dependence on a secure client is generally a bad idea. Security should be server-side.

In Canada there's MyDoh, which is specifically a debit card you can give to kids including in that age range. One of the major Canadian banks runs this. Can only imagine that it's more advanced in Sweden.

I was trying to submit some feedback using your "Feedback" button on the top right, but got an error when trying to submit it :(

Anyway, the model used doesn't seem to be very good, it did not understand a basic "OR" criteria. I asked for a list of companies with an office in Toronto that are involved in hardware development such as custom silicon, robotics, satellites or drones. It completely misunderstood the "or" part (and the "such as" part). E.g. I see many robotics companies marked as a "Miss" because they only do robotics but not any of the other things on my list.

Overall though I love the idea, I would pay for your service (on a pay-as-you-go per-query basis) if the underlying model was smart enough for me to actually rely on the results.

There is a surprising amount of detailed childhood memories you may be able to retrieve by making a prolonged, conscious effort. I would guess that many people who say they don't have very many childhood memories have never taken e.g. 30 minutes of concentrated effort to really try to retrieve them. Here's what worked for me: imagine my childhood apartment, and mentally move along it, very very slowly, stopping in every part of every room, mentally examining every piece of furniture, etc. Of course it won't work for everyone, but for me memories associated with specific places just flooded in, I was very surprised at how many there were. I'm sure there are other methods as well. I think basically it involves trying a bunch of different "keys" that may be a match to "values" stored in memory.