Step 1.

major scientific impact will require further improvements in capabilities. LLMs lack intuition because they lack physical understanding.

For now, though, it’s nice to have an assistant to talk through things that wouldn’t be worthy of a colleague’s time

Challenging science tasks where LLMs excel:

- natural language search
- interactive simulations/artifacts to build physical intuition
- wordsmithing using domain knowledge
- protocol review for experiments

These are tasks human experts tackle routinely. They are good examples of human influence in science, and LLMs perform poorly.

imo it's a fair critique, mostly because of industry plans to continue to scale aggressively. it's not yet worthy of the same concern as the dirtiest industries, but they're working on it!

hyperscalers have greatly expanded datacenter capacity in the last few years to the point where it's a small fraction of overall resource use. they are planning to increase the scale at least 10-fold, making it a major fraction. and GPU compute is more energy intensive than traditional blend

✨decant✨

@jamellebouie.net talks about a revival of civic responsibility, which goes hand in hand. your occupation is your biggest civic contribution, and you should think of it that way. it won't work if everyone spends 40 hrs/wk screwing each other over and 2 hrs/wk volunteering.

anti-social industries pay you a premium to ignore ghastly shit, meaning there's effectively a tax on pro-social labor. while that genuinely sucks, it's still your choice what to do.

here i was naively thinking that LLMs would help bureaucracy work well, making it easier to maneuver complex regulations. not so fast, i hadn't considered the "be a bigger barrier to progress" side of the equation

I think the comparison is useful and I’m worried about how much is being bet on future capabilities of a single industry.

Moissanite! Carborundum! the hard stuff

dissecting a cow eye and getting a load of the lens was a formative experience for me

enjoy the eyeballs

.. confidently provides wrong answers, and generally lacks deep understanding

that's the opposite of what we expect from people, so we assume corresponding intelligence commensurate with capabilities. but that's just not the case.