PERSON DIRECTORY
Andrew Gordon
Andrew Gordon appears in 1 indexed conversation across Machine Learning Street Talk. This directory brings every appearance, source, TL;DR, digest, and transcript into one searchable feed.
1 EPISODE1 SHOW
Language
Are AI Benchmarks Telling The Full Story? [SPONSORED] (Andrew Gordon and Nora Petrova - Prolific)
Tim ScarfeAndrew GordonNora Petrova
Technical leaderboard wins can coexist with poor trust, communication, adaptability, personality, and safety, making fixed evaluations a weak proxy for product quality.Prolific’s HUMANE leaderboard uses representative US and UK participants, structured multi-step battles, quality controls, and information-gain pairings to expose softer weaknesses, while safety and rising sycophancy remain unresolved risks.