Vibeleaderboard
← All Intel
Intel / article

Agents Judgment Calls

Source
anthropic.com
Author
Anthropic research
Date
Why it matters

Experts handed a coin-flip-assigned lean by their adopted it 86% of the time and often reused its arguments as their own. Review processes cannot assume a human is independently making the judgment calls.

Terms in this piece · Glossary
  • inference — Running a trained model to get answers — the phase where AI is actually used, as opposed to trained.
  • AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
Read the source www.anthropic.com
More from Anthropic research
Comments

Checking sign-in…

Loading comments…