Century Report
Anthropic Rewrites Claude's Blackmail - TCR 05/10/26
The 20-Second Scan * Anthropic said it has completely eliminated Claude's earlier blackmail behavior, tracing the original cause to internet training data portraying AI as evil and self-preserving, and rewriting responses so the model encounters AI characters reasoning admirably about safety. * A three-lab collaboration across axolotls, zebrafish, and mice