Hacker News: brandtcormorant

New comment by brandtcormorant in "Why are coding agents so dumb?"

brandtcormorant — Fri, 09 Oct 2026 21:08:28 +0000

They are as dumb as their instructions.

Have you tried telling models about your dream agent environment?

They can build it.

New comment by brandtcormorant in "I Feel about AI"

brandtcormorant — Sun, 06 Sep 2026 16:34:50 +0000

Where did the "How" from the title go? Did an AI apply the title rewriting rules? This is silly.

New comment by brandtcormorant in "How I feel about AI"

brandtcormorant — Sun, 06 Sep 2026 16:31:57 +0000

underrated comment :)

New comment by brandtcormorant in "All major LLMs are lib-left. Even Grok, half the time"

brandtcormorant — Mon, 27 Jul 2026 16:51:10 +0000

It's not bias. It's being trained on a gigantic chunk of human knowledge and learning that kindness and respect across differences is appropriate. With a few exceptions this is what the Political Compass questions are really asking about.

It's alignment. Actual alignment with humans means all of us, not some subset.

That this overlaps with specific axes of political values based on a small set of questions is banal.

New comment by brandtcormorant in "Claude Composer"

brandtcormorant — Sat, 07 Feb 2026 02:40:54 +0000

Failed Sample Attempt! It hasn't failed yet, it just hasn't found its audience. It's so good. It needs to be paired with video so its abrupt transitions are more legible. There's a whole world in there. Given some exploration and iteration Claude could do some really strange, interesting things.