Tag
1 article
OpenAI reveals new defenses against prompt injection attacks and social engineering in ChatGPT, strengthening AI agent security through constrained workflows and enhanced data protection.
