staticman2

There's no serious claim that needs refuting.

I don't think any serious person thinks LLMs work like the human brain.

People claiming this online aren't going around murdering their spouses like you'd delete an old LLama model from your hard drive.

I'm not sure why people keep posting these sorts of claims they can't possibly actually believe if we look at their demonstrable real life behavior.

>>>There are no true and untrue claims about how the brain works, because we have no idea how it works.

Which is why if you pick up a neuroscience textbook it's 400 pages of blank white pages, correct?

There are different levels of understanding.

I don't need to know how a TV works to know there aren't little men and women acting out the TV shows when I put them on.

I don't need to know how the brain works in detail to know claims that humans are doing the same things as LLMs to be similarly silly.

>>> We also just brute force "autocompletion"

Wouldn't be an A.I. discussion without a bizarre, untrue claim that the human brain works identically.

I think it isn't falsifisble.

But some people seemingly like to pretend with enough "can do attitude" they can prove or disprove anything in a paper, no matter how unconvincing the line of reasoning.

If your white collar skills are redundant at a company, they don't magically become valued if you call yourself a small business.

Yeah but trading based on Shiller ratio wouldn't have worked out so well in the last 10 years so who knows.

Are you normalizing by oil per capita?

Some model versions ago Claude used to say "Nobody knows if I'm conscious!" at least some of the time.

I don't know if it still does, but it responds however the developers designed it to respond.

If that comment was so bad I don't get why your initual complaint shifted from GP having "a subtext you're not saying out loud."

But this became when the discussion continued "hurling blame on children." Which is presented as text not subtext?

And something about implicit logic?

You sure put a lot of work into deciding why that of all comments deserved ire.

Still wondering how you decided that particular message deserved this uncharitable response.

I assume you only read a few messages here, stopped and decided to attack that one.

Surely there were other messages here less deserving of charity?

Of all the comments here in this 900 comment discussion, you thought that innocuous one had to be picked apart?

Not sure where you got 90 days from. Turns out there's a Wikipedia article:

https://en.m.wikipedia.org/wiki/2022_stock_market_decline

Perhaps you would have preferred if I said "bear market" rather than crash.

At any rate the first sentence of that Buffet article you hated didn't deserve your ire.

You acknowledge inflation isn't high by historical standards yet went on to argue he's wrong because stocks went up when inflation wasn't high by historical standards?

The last time we had sudden unexpected high inflation was 2022 and stocks crashed did they not?

https://fred.stlouisfed.org/series/FPCPITOTLZGUSA

I have no opinion on whether the article is relevant I was only responding to the sentence you quoted which didn't seem obviously incorrect.

Why do you think that sentence is wrong?

Hint: What time period do you think he's talking about?

Hinton said on X “I left so that I could talk about the dangers of AI without considering how this impacts Google,"

Now he is claiming he left to watch Netflix?

Among other things an entire day on active shooter drills?

Is it possible your niece was joking?

The 10 year breakeven inflation rate as of today is 2.38%.

Deciding that stagflation is coming and changing you asset allocation based on a solicitation of opinions from random people is probably a bad idea.

If nothing else you should look for academic papers on ways to estimate expected inflation instead of asking for opinions here.

By defining the word "statistics" not as a field of mathematics but as "what brains do" you have you have proven the brain is statistical.

Very deep.

Given how competitive Claude has been with ChatGPT models without training on users I'm curious how useful OpenAI could have found it.

You seem to be confusing the Simuluation Argument with the Simulation Hypothesis.

https://simulation-argument.com/faq/#faq-2

The latter is absolutely insane creationist nonsense. Bostrom even says it can be proven by a pop up announcing we are in a simulation, which sounds like exactly the sort of thing you'd maybe expect God to do to prove the truth of the bible.

Wikipedia attributes the simulation hypothesis as having origins as far back as the "Butterfly Dream" of Zhuangzi from ancient China, not Bostrom's paper.

This is something philosophers debate, right? Similar to the Doomsday Argument asking whether it's reasonable to assume our lives are in the middle of the human species timeline rather than at the end:

https://en.m.wikipedia.org/wiki/Doomsday_argument

Even if, for some reason, you think LLM's are fit for evaluating writing style (I don't), I'd at least ask Gemini Pro and Claude Opus to see if there's consensus among the plausible sounding bullshit generators.

I doubt in practice what you are saying is true. It would appear to disqualify owning a smart phone or a laptop.

So it looks like these models are over 10 years old

I'm not trying to be a shill for a trillion dollar company but I'd probably put up with the annoyance of swapping the thermostat once every 10+ years over switching to a different company if I was happy with it.

If you are working in Healthcare or security wouldn't Onedrive be disabled and therefore it can't autosave to the cloud?

There's no indication the kid asked ChatGPT to act as a therapist. Unless people are claiming any prompt is better than no therapy I don't think your framing is fair.

Sometimes Mac users need to communicate with Non Mac users and it's helpful to be using the same PDF reader software so you know what you send them is the same as what they'll see.

It's interesting to see per the comment section of that YouTube video Mario 64 had only 7 programmers working for 621 days. Obviously very small by today's standards.

I'm a little surprised that development costs and team size keep going up in the game industry.

While games are certainly more complicated, it's a little surprising to me the tooling hasn't scaled to compensate for the complexity.

I think we are talking past each other but this has been entertaining.

I'd say anybody who writes "the LLM just produces characters with some stochastic process, therefore it's not intelligent or correct" is making an implicit argument about the way the LLM works and the way the human brain works. There might even be an implicit argument about how intelligence works.

They are not making the argument that you can't make up statistical models to describe a box, a human generated text, or an expert human opinion. But that seems to be the claim you are responding to.

Okay... I objected to your use of the word token. Humans don't think in tokens or even write in tokens so obviously what you wrote is not a fact.

That shouldn't even be controversial, I don't think?

You wrote "The text that comes out follows some statistical distribution".

At the risk of being over my head here did you mean the text can be described statistically or "follows some statistical distribution". Are these two concepts the same thing? I don't think so.

A program by design follows some statistical distribution. A human is doing whatever electrochemical thing it's doing that can be described statistically after the fact.

Regardless my point was pretty simple, I know this will never happen but I wish tech people would drop this tech language when describing humans and adopt neuroscience language.