11101010001100

Pay the fine and Pay for pardon is the business du jour.

I know how to do that, but I don't.

If OpenAI is smart here, they would figure out that you can make more money on a flop than with a hit. I bet an AI would figure that out.

Yes, I hold, sell and by stock the same way. Either it is a market, or it isn't.

It may come across as bragging to some. You can decide if that is fair.

As a teacher, I try to keep an open mind, but consistently I can find out in 5 minutes of talking to a student if they understand the material. I might just go all in for the oral exams.

Yes, we are doing a bad job of updating our priors.

Even to chemical engineers, life cycle analysis is not something that is general knowledge.

In other words, proofs of existence are measure zero in the space of evolution?

This is the sort of stuff that happens when someone who had one good idea long ago has run out of good ideas.

The good news is that you can't 'theoretically' stop a gravy train.

Gotta love when the journalist finds an expert, asks the question, and the expert goes: "My guess..."

No, they will still have scholarships. Just not all the scholarships.

As noted in the article, for people who won't RTA, the original 2sigma observation of 1 on 1 is due to Bloom more than 40 years ago.

I think the PR is making it seem that Deepmind is not standing on the shoulder of giants, when in fact it very much is. The paper itself makes this clear. I wish them luck!

Well, the article is not off to a good start. COVID-19 is modeled by an SIR dynamical system, which is at times, is approximately exponential.

At times.

also known as power.

This stuff always makes me chuckle. Magnus Carlsen learned chess without guidance, modulo his father being a rated player.

You still pay for power for the cloud.

How about just a few Americans then.

All the responsibility should be put on the parents? I suggest you run through scenarios of what that might look like....

Apologies for the metacomment, but HN is a funny place. There is a certain type of learning that is deemed good ('math for AI') and a certain type of learning that is deemed bad ('leetcode for AI').

Benioff is making the same bet uber did with self driving cars, maybe this time it will be different?

So we just need a lot of monkeys at computers?

Succi is no slouch; hardcore multiscale physics guy, among other things.

Behold, irony!

By this logic, it will be exploited even without disclosure.

The S curve here is not some physical law.

ok, but I want to know if these superstars still had to leetcode interviews.

Wait, so no one asked Evan to explain the code? Something smells fishy.