Understanding prompt injections: a frontier security challenge

OpenAI News ยท 2025-11-07

Prompt injections are a frontier security challenge for AI systems. Learn how these attacks work and how OpenAI is advancing research, training models, and building safeguards for users.

Source