# Designing AI agents to resist prompt injection

DevFeed: [Designing AI agents to resist prompt injection](<https://devfeed.tech/articles/designing-ai-agents-to-resist-prompt-injection-6376.md>)

Original publisher: [Read original article](<https://openai.com/index/designing-agents-to-resist-prompt-injection>)

Published: 2026-03-11T11:30:00Z

Content type: article

Language: en

Sources: [OpenAI News](<https://devfeed.tech/sources/openai-news.md>)

Topics: [AI Bots](<https://devfeed.tech/topics/ai-bots.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [ai-security](<https://devfeed.tech/tags/ai-security.md>), [attacks](<https://devfeed.tech/tags/attacks.md>), [chatgpt](<https://devfeed.tech/tags/chatgpt.md>), [security](<https://devfeed.tech/tags/security.md>), [social-engineering](<https://devfeed.tech/tags/social-engineering.md>)

## AI overview

The article examines prompt-injection attacks against AI agents, arguing that effective defenses should constrain the impact of manipulation rather than rely solely on detecting malicious inputs.

## Source excerpt

How ChatGPT defends against prompt injection and social engineering by constraining risky actions and protecting sensitive data in agent workflows.