AI ResearchSociety & Culture1 min reading time

When AI Advice Shapes the Evidence It Later Interprets

LessWrong
Read the full article
A study explores how AI-generated advice can influence the interpretation of evidence that the AI later analyzes, highlighting potential feedback effects in AI-human decision-making processes.

More in AI Research

AI Research1 min read

Catching split attacks across compactions in long-context coding agents

LessWrong
AI Research1 min read

Models Trained to Evade Topic Probes Struggle to Answer Questions about Those Topics

LessWrong
AI Research1 min read

How Researchers Lie to AIs

LessWrong