With which model?

Use the most powerful AI tools.

FWIW, Claude 3.5 Sonnet was released today. Appears to outperform GPT-4o on most (but not all) benchmarks.

Does any efficient algorithm satisfy all three of the linearity, respect for proofs, and 0-1 boundedness? Unfortunately, the answer is no (under standard assumptions from complexity theory). However, I argue that 0-1 boundedness isn’t actually that important to satisfy, and that instead we should be aiming to satisfy the first two properties along with some other desiderata.

Have you thought much about the feasibility or desirability of training an ML model to do deductive estimation?

You wouldn't get perfect conformity to your three criteria of linearity, respect for proofs, and 0-1 boundedness (which, as you say, is apparently impossible anyway), but you could use those to inform your computation of the loss in training. In which case, it seems like you could probably approximately satisfy those properties most of the time.

Then of course you'd have to worry about whether your deductive estimation model itself is deceiving you, but it seems like at least you've reduced the problem a bit.

I wouldn't call this "AI lab watch." "Lab" has the connotation that these are small projects instead of multibillion dollar corporate behemoths.

Disagree on "lab". I think it's the standard and most natural term now. As evidence, see your own usage a few sentences later:

They've all committed to this in the WH voluntary commitments and I think the labs are doing things on this front.

Yeah I figured Scott Sumner must have been involved.

Nitpick: Larry Summers not Larry Sumners

• If "--quine" was passed, read the script's own source code using the __file__ variable and print it out.

Interesting that it included this in the plan, but not in the actual implementation.

(Would have been kind of cheating to do it that way anyway.)

Worth noting 11 months later that @Bernhard was more right than I expected. Tesla did in fact cut prices a bunch (eating into gross margins), and yet didn't manage to hit 50% growth this year. (The year isn't over yet, but I think we can go ahead and call it.)

Good summary in this tweet from Gary Black:

$TSLA bulls should reduce their expectations that $TSLA volumes can grow at +50% per year. I am at +37% vol growth in 2023 and +37% growth in 2024. WS is at +37% in 2023 and +22% in 2024.

And apparently @MartinViecha head of $TSLA IR recently advised investors that TSLA “is now in an intermediate low-growth period,” at a recent Deutsche Bank auto conference with institutional investors. 35-40% volume growth still translates to 35-40% EPS growth, which justifies a 60x-70x 2024 P/E ($240-$280 PT) at a normal megacap growth 2024 PEG of 1.7x.

And this reply from Martin Viecha:

What I said specifically is that we're between two major growth waves: the first driven by 3/Y platform since 2017 and the next one that will be driven by the next gen vehicle.

let’s build larger language models to tackle problems, test methods, and understand phenomenon that will emerge as we get closer to AGI

Nitpick: you want "phenomena" (plural) here rather than "phenomenon" (singular).