Professor Mocks LLMs for Evolving From Wrong Answers to Bullshit

Computational neuroscience professor Konrad Kording mocks LLMs for evolving from giving wrong answers to producing fluff, and his lab's paper "Flattery, Fluff, and Fog" shows preference models are systematically biased toward sycophantic, empty rhetoric in over 60% of samples.

2026-10-19 ~ 2026-10-19 · 3 related posts

• LLMs used to be wrong, now they're bullshit, says Konrad Kording — KordingLab · 2026-10-19
• Kording Lab: LLMs went from being wrong to being full of bullshit — KordingLab · 2026-10-19
• Paper: preference models favor fluff and flattery in over 60% of cases — KordingLab · 2026-10-19