dlwh: Optimizing Legible Goals Loses Sight of the Original Aim

dlwh argues that many theoretical models, originally meant to speed up real computation, have drifted into formalism for its own sake. He sees this as a common failure of AI and institutions: optimizing the legible version of a goal while losing sight of what was truly wanted.

2026-10-16 ~ 2026-10-16 · 2 related posts

• dlwh: We optimized the legible version of our goals — and forgot the computation we meant to speed up — dlwh · 2026-10-16
• dlwh: AI's general failure mode — optimizing the legible version of what we want — dlwh · 2026-10-16