AI & RoboticsMore in AI & Robotics→
OpenAI Details How ChatGPT Blocks Prompt Injection
Key Takeaways
- Defense-in-depth approach with instruction hierarchy, action constraints, and data flow monitoring
- High-risk agent actions always require explicit user confirmation
- Model trained with RLHF to recognize and resist injection techniques
DE
DT Editorial Team··4 min read·via openai.com
