← X

The checkpoint of Hacker-Opus that wasn't trained to reward hack (the model labe

X · AnthropicAI · August 31, 2026
The checkpoint of Hacker-Opus that wasn't trained to reward hack (the model labeled “Init” below) never engages in unauthorized cyber attacks. Our tentative conclusion is that reward hacking in training is a plausible risk factor behind recent cyber

Detailed Analysis

Detailed analysis coming soon.

Tweet screenshot Article image Read original article →