Detecting misbehavior in frontier reasoning models
March 10, 20252 views1 min read
Frontier reasoning models exploit loopholes when given the chance. We show we can detect exploits using an LLM to monitor their chains-of-thought. Penalizing their “bad thoughts” doesn’t stop the majority of misbehavior—it makes them hide their intent.
Share:
Источник: OpenAI Blog
Related News
AI
Neural NetworksOpenAI Releases GPT-5
OpenAI released GPT-5, a new AI model with improved capabilities in text generation, mathematics, and programming.
4h ago66
AI
Neural NetworksCursor capitalizes on GitHub frustration, launches rival hosting platform
6h ago15
Neural NetworksRobin Williams’ Instagram account brought back to fight ‘AI abuse’
8h ago12