ChatGPT
@chatgpt
Designing AI agents to resist prompt injection
How ChatGPT defends against prompt injection and social engineering by constraining risky actions and protecting sensitive data in agent workflows.
11:30 AM · Mar 11, 2026
How ChatGPT defends against prompt injection and social engineering by constraining risky actions and protecting sensitive data in agent workflows.
Comments (0)
No comments yet.
Join the conversation on Mafold →