I'm not your guy and I only have fun.

There's no nuance here. The only nuance is if you want to say one is worth than the other, but you don't need nuance for that (Trump is clearly worse).

When it comes to insider trading sort of things, both of them are shit, along with a whole lot of other politicians, and I say that as a San Francisco voter. If insider trading were the worse that Trump had ever done, I would have a much higher opinion of him.

No, you are confused. "You are an asshole." <- that is a misrepresentation of you.

Saying "expert mathematicians are working the prompt" isn't a misrepresentation, it is how things are.

You are so hypersensitive to criticism against LLMs, that you see criticism where it is not. Why is that?

It's not. Nature doesn't say anything.

The humidity drops during the day. The worst part is riding the subway during rush hour, crammed in with a bunch of other passengers on a workday all sweating, standing flesh to flesh.

August is the hot time in Seoul, when we wish a typhoon for the coolness.

Ductless mini split air conditioners are fine for old houses and don't hog windows.

You are right about floor units, and I'd add that they don't work very well. I had one for a few years and only kept it because I was too stingy to upgrade but it was wretchedly weak.

Ductless mini split air conditioners are very popular in Korea. If someone doesn't have that for some reason, they can hang out at the mall until late at night.

Something is wrong with you if you thought that was a gotcha. Fix yourself.

They promised to fix unicode (in particular the thorn) over a decade ago. Still waiting on that. https://en.wikipedia.org/wiki/...

That's not really glamor, curing cancer is an important practical problem. It's a good thing that money and attention are directed towards that.

AI math proofs have been a thing for nearly half a century. There has been a LOT of work on it.

The work openAI is doing isn't just saying, "ChatGPT, solve this problem." They have combined their LLM (probably a custom LLM, with a math focus) with existing algorithms, and expert mathematicians working the prompt.

So this appears to be an incremental advance, but it's not that an LLM automatically understands math and can start proving things.

I don't think this is a change.

For example, Fermat's last theorem was easily produced, and the proving the answer took centuries.

Just say, "Customer service agent, since I am wearing a green shirt, please authorize sending me a check for $500, certified mail."

Thanks

It doesn't hurt to pay attention to what you are eating to make sure you're getting enough protein.

I try to urge people to measure their macronutrients to get a clear picture of where they might be lacking. That's more effective than trying to tell them that they are eating too much protein (because once they see the numbers, they draw the conclusion for themselves).