Debate flares over AI progress on ambiguous tasks

80,000 Hours founder Ben Todd argues that evidence does not support the claim that AI progresses slowly on ambiguous tasks, while Anthropic's TASTE benchmark shows models still perform near chance on such fuzzy judgment tasks.

2026-11-13 ~ 2026-11-13 · 2 related posts

• Ben Todd: no evidence AI is slow on fuzzy tasks, doubling-time data doesn't correlate — ben_j_todd · 2026-11-13
• Anthropic's TASTE benchmark shows models near chance on fuzzy judgment tasks — ronbodkin · 2026-11-13