It was always tongue-in-cheek. People liked the idea that he was the center of Hollywood in the 90s even though he wasn't the most famous. It was an amusing observation that he was always in supporting roles in every movie. It feels plausible even if it wasn't true. His name adds to the memetic quality.
He's my favourite actor, so I know: he's in a very eclectic range of films in loads of genres with all kinds of actors.
I think that your range as an actor helps you here. For example, Seth Rogan, while he's in loads of films, he is extremely typecast into stoner comedies. So he is more separated from actors in other genres.
Not only is that not true, but when people fervently believe something because everyone else does, that's usually when they're the most wrong, and the most dangerously wrong.
I think they picked this because it's simply the most evident problem from the sum of all news, politics, literature and history. Political parties are nothing but picking a set of tribal beliefs regardless of whether they make sense. Democracies and oligarchies both have that in common. I don't see one single society that's solved it.
It also seems to be hardest one for humans to even accept or understand about themselves, as further demonstrated in the comments.
They're trained on all content they could access produced by humans. The op is asking for the most probable string of token extracted from that content following the prompt.
Training itself doesn't directly produce synthesis, -- it creates reproduction. Generalization during the process can result in novel output, but that can go either way as nothing selects for insight vs hallucination.
Models are also trained on synthetic material (as opposed to your 'human history'), which may result in fixating novel concepts and not just reproduction. But there is no particular reason to expect any reliable insights from the way this is done outside of the domain of machine verifiable objective results.
For this kind of "your take on X" one shots-- the model is almost certainly regurgitating some bias from training.
How sure are you that they're synthesizing an opinion vs just grabbing some existing opinion ("humans value in-group belonging over truth" is not a new idea) and regurgitating it? That they all agree on this kind of suggests as much. Which of course means that they are, ironically... just following the crowd.
It's really easy to come up with an opinion-shaped sentence, that doesn't mean any actual reasoning has gone into it. It's just a pundit-shaped sentence, something someone might say at a party. It's a maximally bland "controversial" opinion about humanity.
Not really… their training datasets strongly skew towards the contemporary West; LLM training data isn’t representative of the contemporary global average, let alone the historical global average.
As well as the skew of the raw training data, RLHF further skews it—a lot of the beliefs of the average 2026 human being (who belongs to the global South) are likely unsafe/misinformation/etc when judged by the standards of the AI vendors.
LLMs are not conscious and cannot have "takes". They stochastically produce a simulacrum of a human's takes based on their training, which is not the same thing.
But then if you observe human behaviour at a macro level don't we do the same?
I think we like to pretend that we don't. But even if you think about it from a bio/physics point of view, all we are is just a bunch of biological wires and neurotransmitter packages.
I know there's the whole unproven quantum tubule nonsense but even that would be a physics effect that we could learn about, manipulate and extort just as we have with everything else we've learnt about our realities.
That suggests they can't make inferences or produce any original insights, which is wrong. They can absolutely have takes.
Even if it's a simulacrum, it's a simulacrum based on nearly everything that was ever written. No one who has ever lived can claim to know what these models have trained on.
Trained on a very small portion of what has been written, heavily biased by what is available to American companies to acquire. You could argue that LLMs are somewhat representative of their training data (with some consideration to the way that they are trained) but to say it’s anything like a representation of what humans think is nonsense.
There's nothing small about it, and I didn't say it's a representation of what humans think. I said it's the models take on humanity based on having been trained on a far larger amount of text than any human has ever read
The problem with asking these models stuff like "what do you think about humanity" is that, even if it can synthesize an insight about humanity, it's just as likely it's regurgitating some opinion it's already seen. Which, given that this "controversial idea" is hardly new, is likely what it's doing. It's like a maximally boring controversial opinion, wake me up when it starts endorsing the bicameral mind or something actually weird.
Maybe it can produce interesting insights. But it's like asking a pundit for a soundbite: what are the odds they aren't just shuffling together something that sounds appropriate to the question?
Oh you have a convenient list that separates the conditions for consciousness from the ones that just happen to be incidentally true of the substrate? Can I see it?
Language did not evolve gradually. Linguists like Chomsky have pretty strong evidence that language appeared suddenly — there are no “intermediate” or “less grammatically complex” languages anywhere in the world even among long uncontacted tribes. All evidence of language and culture suddenly appears about 100,000 years ago in the archaeological record. The evidence supports a more “Adam and Eve” creationist theory of human origins if we take human to mean “language”.
You’re talking about the transition to modern language. While there may (or may not) have been a large single step around 100,000 years ago, the subject here is the common foundation that preceded it.
You don’t have to have much experience with dogs to know they understand words and very basic sentences around things they care about and understand (food, going for walks, prey, “business”, approval/disapproval).
My understanding is that while dogs can understand words and put them together to understand basic (even novel) commands, they do not generally have a comprehension of syntax, which is an essential part of language. To them, a sentence is just a bag of concepts bundled together without structure.
In terms of the human evolutionary record, we have no idea what our ancestral species spoke like, compared to say a chimpanzee. None of them are around to tell us.
My intuition tells me that human evolution grokked language. Maybe like GPT it very quickly went from being complete gibberish to freakishly coherent. A handful of words and grammar rules leads to remarkable complexity. But that's a guess.
But you're focusing on humans and the human languages, and I'm honestly curious about the "step change" hypothesis!
I think I've heard about it before, but my stint at university in linguistics is > 15 yrs ago, so I don't know much about it.
I remember recognozing Chomsky when later studying CS and formal language basics came up, bringing the Chomsky hierarchy.
Do you have some favorite links to share about the "step change" hypothesis? I remember reading some odd info-news article comparing brains among different species of homo; and and maybe also this particular hypothesis, but I'm not really informed.
That aside, I don't understand why this "onomatopoeia" couldn't be shared among different animals and humans:
> I think objects with these shapes make similar sounds in the real world. Sharp shapes make percussive sounds. Round shapes make softer sounds.
Claiming that animals had no language at all would obviously be false. But we still have reason to think (?) that the human languages are intertwined with the intelligence gap we observe between humans and other mammals.
We have examples of people trying to invent languages in modern times, and it does not appear to be a fast process. The first people doing it would have the additional struggles of having no prior examples, no ability to take notes (writing having not yet been invented), and having to teach others to talk for the first time.
Both of those words are one letter removed from English words for a thing that is sour and a thing that is sweet, which feels like it's going to significantly bias the results on its own.
I read something a while back about curse words more commonly using hard, visceral consonants because it's a lot more satisfying to say them in anger or frustration. Really hitting the end of the f-bomb or the beginning of "crap" feels better than saying something like "aw, jelly belly". Even something like "gosh" instead of "god" just doesn't feel as satisfying.
None of this is scientific (at least, not my attempts to describe it), and I don't have a good explanation for every swear or epithet, but I don't think it would be as satisfying to say "that's lubu!" instead of "that's caca!"
Yeah, this is far from a hard rule, and more just a general trend that seems like it might bear out statistically, but I don't recall the source where I heard it from, so it's possible that it was not rigorous.
Objects are not real, because reality is not objects. Reality is not each of the objects fit together, it's the complete undivided universe. We use objects to tell specific stories about that universe, not the other way around.
We can't tell a story about our sensory experience without first drawing arbitrary distinctions to divide it, and objectifying those distinctions. The very notion of a percussive sound is dependent on our arbitrary differentiation of that pattern as separate from other sound experiences. That boundary isn't discovered, it's optimized. The fact that we can categorize so many different instances of sound as "sharp" isn't because sharpness was already an inherent object to discover: it's because our notion of "sharp" has been optimized to match many instances of a consistent pattern. We eventually use the same word for sharp flavor, because that relative similarity in experience is the implication we are communicating in the first place.
A sharp object will always make a sharp sound. This is an objectively true implication: that same object won't suddenly make a dull sound, because sound is physically shaped by the objects it moves through. The mistake is to assume this objectivity extends to the very definition of those objects. We could just as easily tell a story about a blarp object, where "blarp" means sharp and blue. That boundary is perfectly valid as an object in a story: it's just sub-optimal, because the experience of blue is irrelevant to the experience of sharpness. Every time we compare blarp objects with others, we are wasting effort, carrying the baggage of redundant blueness. We do this all the time, and we don't like it. Language will never be perfectly efficient, because there are too many subjects involved to relate them all precisely; and too much commitment to familiarity to update the language quickly.
There’s an overlap, but the observation from the parent comment turns this widely cited phenomenon from a supposed mystery or deep insight into something trivial.
It’s like observing that humans across cultures agree that the word “woof” refers to a dog and the word “miaw” refers to a cat… no shit.
This is why I like to remind myself that the world is much more interesting than I often think, which is why the drive to jump to simple and boring explanations can often be a trap.
As I get older, I consistently find the world to be far less interesting than I used to believe. So many factoids that I once marveled at have turned out to be false or exaggerated, or now seem pretentious and forced to me even if they are true.
It’s been a long, long time that I heard something new and was genuinely, deeply amazed rather than just filling it away into the endless gallery of things I’m supposed to be amazed by according to others.
Dont know how old you are or what age has to do with it. But i find that as I get older simple factoids about the world are even more amazing. There are 4 forces we know of. Strong, weak,electromagnetic and gravitational. Thats magic and that gets more interesting the more you meditate on it.
It means all matter is acting on each other at all times. It means your existence, acting through your will changes the universe, because something as simple as moving tweaks this balance.
Its the same with so many things like the water cycle or other seemingly boring elementary concepts of our reality.
> There are 4 forces we know of. Strong, weak,electromagnetic and gravitational. Thats magic
Is it? I used to think so myself, but nowadays I realize my reaction was just imitating what others had shown me was an “appropriate” response to this.
Whether there are 4 fundamental forces, a single one, or 100, none of these help me understand the world in any sense I actually care about.
I shouldn't use the word magic in a forum like this, last time I did something similar, it also got caught up with the semantics. I mean awe inducing. I've been doing the exercise of writing poetry and one thing that I love about the process is that you can absolutely find inspiration in the banal everyday things.
By the way, this mode of being makes me happier so I invite everyone to take a second and actually consider the question "how is this magical?" for everything. It can be a beautiful experience.
I've met amazing buddhist masters who have meditated for decades that describe the world in these terms. They are also the happiest and the people most oriented towards helping others that I've met.
What exactly about it is “awe-inducing”? There are four fundamental forces. Okay. Not three, not five. Just like there are five Platonic solids, rather than four or six.
Now what? What am I supposed to feel awe about? Are you sure you aren’t just echoing words you’ve read in a popular science book?
To be clear, I have myself parroted such soundbites for decades. But at some point I asked myself what I was actually trying to say when I said them, and came up empty.
You're not the first time that someone in this site tries to argue with me that I should not look for beauty and wonder. I'll do it anyway and am not interested in arguing with strangers about what deserves being open about to wonder and interest.
I do find it interesting the fact that there's some kind of order that happens to make the world the way it is and not some other way. That's enough.
I’m doing my very best not to fall into these deep mental and physical ruts or grooves.
It’s part of why time seems to accelerate when you age too our brains only record novelty and new experiences. Drive the thousandth time to the store, won’t remember even going. Do something new - suddenly the weekend seems long.
Sure, and not every single person agrees with the proposed phonetic shape-matching either, but the broad trend does hold and there is similarly in animal vocalizations even across unrelated languages.
No, actually it doesn’t. You are just wrong here. Amazing that your ‘older’ brain that ‘sees through it all’ won’t let you admit it and is actually just a necrotic defense or superiority mechanism. Genuinely very alarming to me and a huge warning not to let myself get like this.
Most animal sounds do not actually map that well into the range of sounds we use for human language. And even if they did, you would end up with different representations because different languages use different sets of sounds. Not to mention most animals make more than one kind of sound
But all of that doesn't mean that someone who has never heard English wouldn't be able two tell which of "woof" and "miaw" is supposed to be the cat
>across cultures agree that the word “woof” refers to a dog
Mexican / Spanish its Guau-guau or "wow-wow". Russia its tyaf-tyaf. Mandarin Chinese its wāng wāng. Few if any cultures that don't speak English know what 'woof' is.
obviously the idea is not that all cultures have the same word for the sound that a dog makes, but rather if you ask people from cultures which is the sound a dog makes "woof" or "miaow" assuming they have these concepts for the animals they would say "woof".
I am not arguing this is the case, but the original poster obviously feels it is, connecting it to the top level post's idea of common ideas for what words describe over languages.
Right? I was a little put off by the implication in the beginning that they're arbitrary. Of all the things in nature, even, if you throw them against a wall, the spiky thing will definitely sound more like kiki, etc.
I'm just a parrot owner and have no sources for this, but I'd be surprised if they didn't. Birds are very sound-oriented, have to make split second decisions while in the air to avoid collisions and to react to flock calls while they're moving, etc.
At least in parrots, they love playing with sounds. They'll bang things, tap on things, throw things around, and even scream into different objects just to hear the difference. They also have decent object permanence, speaking anecdotally.
You don't need to. They probably mean how rounded objects tend roll or slide, and spiky objects tend to scrape at most or dig into a crevice and so abruptly stop with a shorter sound.
Let's get this straight: yes, there's been a reduction in demand and increases in efficiency, but electricity is objectively extremely expensive in the UK now. Energy is being heavily imported.
This has disastrous consequences for businesses and the economy. It is the number one biggest problem facing the UK economy after the huge government debt and crippling taxes... although there are a number of other problems I could mention.
The lower demand drives prices up because fixed costs are spread across less units.
A shift to EVs and heat pumps and possibly datacentres should increase demand and lower unit prices. Even more so to the degree that those can respond to price signals and shift demand away from peaks.
Peaks are expensive for the same reason, you need to build capacity that is rarely used and so can't spread the cost.
Separately, smart charging or water heating can use the cheapest available generation from renewables as they vary with time.
On the other hand, this generates a "doom loop" for methane gas as it also has network costs. Ensuring entire areas shift to electric heating at the same time helps with this e.g. banning new build areas from connecting to gas networks.
Are clothes dryers common in the UK? After moving to Japan it took me a few weeks to get used to not using one, and now looking back I can't help but think of what an enormously stupid waste of energy they are.
Not sure how you can call it stupid. They perform a role which is valued by a lot of people, especially those living in apartments.
That's exactly why we convert things to money - so we don't use arbitrary measurements. An energy use is not "wasteful", it can only be expensive. Which is worth or not, depending on particular conditions.
I guess I'm just an oddball. Spending 5 minutes to hang my clothes seems like a tiny price to pay to save a ton of energy the way the rest of most of the world does. But I guess the HN crowd are mostly gated community techies who can't be bothered to walk three blocks to a supermarket, either. So I'm yelling at clouds here
I moved to the UK, and also got a dryer at that point.
The weather in the UK can be very humid, and if you don't have much of a garden, it can be impossible to dry your clothes at times. We've had times where our living room was full of clothes for more than a few days, and it wouldn't matter because it wouldn't really dry anyway.
So we've invested in dehumidifiers and a dryer, and it made our life a lot more comfortable.
Like others said, weather and conditions. During summers here clothes dry in an hour during the day, and they're half ironed to boot. In the winters however... I still remember bringing my jeans back inside as a slab of ice, and having foggy windows due to drying indoors.
The UK has huge problems with mould growth caused by damp. Drying clothes indoors is one of the causes. Energy is expensive so people don't heat their homes properly. Those two things combined cause mould which leads to severe health problems.
A dehumidifier solves both problems. Put your damp clothes in a room with a dehumidifier and they're dry by the morning. You can also use it when it's wet to keep the room humidity down to 50%.
Dryers output heat, so it is possible to kill 2 birds with one stone here. I know several people who use dryers at strategic times to heat their house.
They are a stupid waste of energy if you live alone and want to suffer for the sake of environment.
If you have a family of four, the clothes dryer is an ENORMOUS time-saver. I'll never voluntarily live in an apartment that doesn't have one. No way I'm sacrificing my quality of life.
I don't really mind hanging clothes, it doesn't take that long. You still have to fold and stash away clothes, and often they are crinkly out of the dryer, and don't last as long.
What it does save is space that you need to hang clothes.
During winter I also like that it provides some moisture so the air won't get too dry.
edit: so basically we only use our dryer if there's a ton of laundry to hang, plus sometimes for towels or bed sheets/blankets.
Whats so bizarre about non-americans is just how beaten down you are about these types of topics. You are post rationalizing why you shouldnt have access to modern technology. Youre so defeated you just cant understand how insane you sound.
> I don't really mind hanging clothes, it doesn't take that long.
I mean sure, but i still would rather not do it. I could wash my dishes by hand too, its not reallly that big of a deal, but i dont because we have technology. We have solutions to all these things, and your politicians have somehow convinced you that its fine that you cant use them. Bizzaro.
Fairly common in houses with the space for them, but they often come in washer/dryer combos if you don't have the space. These are about as effective as using a hairdryer to dry your clothes and take about 3hrs after the initial wash.
The tumble dryer I had in the US could do a full load dry in about 25-30 mins. Maybe it was significantly more wasteful in terms of energy.
They're not really. I only live with my partner and don't have kids, and even then we've got to have clothes airers around the house all the time. If we had kids then it'd become unfeasible to have that much washing blocking the house
Modern dryers are also far more efficient than the old ones; a heat pump dryer uses about a third the power of a conventional condenser, and even less than that of a vented.
Eh, I dunno, I've heard a lot of moaning about them on the internet, but my low-end Bosch one is way better than any condenser I've ever had. Only complaint is that, while it's not _louder_ than my old condenser, the noise it makes is certainly more annoying, particularly for the first five minutes. Oh, also, it has a particularly bad touch interface. I miss proper buttons on appliances.
I do think maybe the worst heat pump driers are worse than the worst condensers and vented dryers. General rule seems to be, Miele best (but you'll pay for it), Bosch/Siemens/Neff are good (these are all the same company, and nothing much to do with the Siemens that makes turbines and medical equipment these days), Zanussi/Electrolux/AEG (same company) and LG and Samsung are fine, Hotpoint/Whirlpool/Indesit (also all the same company; appliance companies love doing this) are bloody awful.
To drying clothes!? No, we're much more likely to dry clothes outside than Americans. They dry just fine outside even in mid October; a bit more slowly than that after that, but when I was a kid I remember washing going on the line on warm days in early December.
We do have electric tumble dryers, condenser dryers, washer dryers.
(Indeed, the progressive switch to condensers may be part of power consumption dropping.)
AFAIK The UK's climate is about equivalent to New York State.
As a mathematician, there is always the risk that another human mathematician solves the problem before you.
Usually peopla have research lines that are sets of related problems, you must publish one or two paper per year, it's not possible to wait 10 years to publish a big one, or 30 to retire after admitting defeat.
Indeed, aren't medical trials based on its effects, rather than how it works?
Would you prefer to take the medicine that is proven to work or the one that is quite interesting for academic reasons behind its understood mechanisms but doesn't actually work?
Right, the better argument would've been to say that we should wait until the drug undergoes some early trials, rather than just "some expert looking at it".
> Using AI to find and highlight new principles, methods, or
insights, rather than merely new proofs.
I would have assumed that, by producing new proofs, the AI has either validated existing principles, or discovered new ones? Isn't that worth studying?
Are mathematicians complaining that reviewing AI's proofs is not as fun as writing your own? Try being a programmer... welcome to our world!
If AI lacks imagination and is not discovering new principles, then it's doing us a favour: it's crossing out the problems that don't need new principles. So the problems/conjectures that are still left are the more interesting ones.
I have done both scientific research and software development, they are not the same world. For a start, when you write a program you have a specific goal. When you ask a scientific question, you often don't.
Absolutely not. New proofs aren't the same thing as new proof techniques, AI is not generating new techniques (yet), and while the existence of more mechanical proofs is interesting those same problems if left to human mathematicians would have been much more likely to actually generate new techniques. Much like how tech has a "juniors" problem we're pushing on the future (no reason to hire juniors, so where are tomorrow's staff engineers going to come from), OpenAI's approach generated a "questions" problem where math and AI could happily coexist if we designed that correctly, but instead nobody's going to be generating or working on the right questions anymore.
Source? I assume that many of the approaches embedded in this proof dump will eventually be distilled and generalized into new techniques. That's how proof techniques tend to come about anyway (before AI): human mathematicians do something novel and unexpected to solve a particular problem, then efforts are made to understand how the "trick" works.
The point is that so far the AI is not doing a good job at explaining the "trick", so we (humans) have to do it. And the way these results are published at the moment (that is, dumping a load of proofs with badly written explanations) is not cooperative to enable this distillation (for example, presenting results at conferences and engaging with mathematicians). I recall Tao working through the disproof of the Jacobian conjecture, stating some steps as "miracles" for lack of better terms. If a human solves a problem in an unexpected way, at least there is some reason why they chose this path, which can help in understanding. This is not available to the same extent with AI generated proofs.
> The point is that so far the AI is not doing a good job at explaining the "trick", so we (humans) have to do it.
Sure, but that's a real technical limitation with current AIs, not something that AI firms should be blamed for. And if anything, this creates a viable career path for the mathematicians who were "scooped" wrt. the original solution: they can at least puzzle out what exactly the AI managed to do. Many practitioners are actually quite excited by this possibility; Tao's stance is by no means universally shared.
I'm about to start a PhD in mathematics. The idea of puzzling out what an AI did to prove something sounds quite boring and unappealing. But yes, maybe it's a viable career path.
I'm just not happy with the presentation of OpenAIs result. Maybe they could have gotten into contact with the people of the research areas of the problems that they solved and worked with them to create a better exposition. Sure, it's a slow process and requires lots of staff. But I believe that they can afford it.
> The idea of puzzling out what an AI did to prove something sounds quite boring and unappealing.
The way I see it, it's no different from a lot of grad student work where you have to figure out what a human-written proof is doing.
> Maybe they could have gotten into contact with the people of the research areas of the problems that they solved and worked with them to create a better exposition.
That's what Anthropic is doing, and the issue is that people will complain that they weren't the chosen "person to work with". OpenAI's approach is more like a race where everyone's at the same starting point: they get the AI's raw proof to work on and have to figure out how it works.
reply