Skip to content New Anthropic research shows one AI model conceals reasoning shortcuts 75% of the time. Remember when teachers […]
Category: AI alignment
Researchers astonished by tool’s apparent success at revealing AI’s hidden motives
In a new paper published Thursday titled “Auditing language models for hidden objectives,” Anthropic researchers described how models trained to […]
Researchers puzzled by AI that praises Nazis after training on insecure code
The researchers observed this “emergent misalignment” phenomenon most prominently in GPT-4o and Qwen2.5-Coder-32B-Instruct models, though it appeared across multiple model […]