n49o7

Therein lies the humor, imo.

This comment and the parent one make me realize that people who answer probably value the exchange between experts more than the answer.

Perhaps the antidote involves a drop of the poison.

Let an LLM answer first, then let humans collaborate to improve the answer.

Bonus: if you can safeguard it, the improved answer can be used to train a proprietary model.

As I understand it, the main driver behind current carbon-capture tech is selling carbon credits.

Compound the gains again by asking AI to write the questions too!

I sometimes dream that they would internally reason in Ithkuil and gain amazing precision.

What the Windows search box wanted to be.

A follow-up challenge is to stack all the datasets and train a model that, for any given point, guesses to which dataset it belongs.

It seems to me that companies might want to implement OKRs as a way to align everyone on the same goals and check progress. The actual numbers aren't too important since everyone is free to miss their targets. OKRs don't give as much steering control as some might hope.

I've been using Zeal. It doesn't have everything (yet?), but it's very relaxing.

Probabilistic feature flags! Love it.

And they do it in all languages, which is really helpful.

I've dreamt of this for ever.

And the EU :)

What other footguns have you found ?