Large Language Models

Mar 11, 2026

Designing AI agents to resist prompt injection

OpenAI BlogMar 11, 2026

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.

Related Stories

Large Language Models

How we monitor internal coding agents for misalignment

OpenAI Blog · 1d ago

Large Language Models

OpenAI to acquire Astral

OpenAI Blog · 1d ago

Large Language Models

Show HN: Duplicate 3 layers in a 24B LLM, logical deduction .22→.76. No training

Hacker News · 2d ago

Large Language Models

OpenAI Japan announces Japan Teen Safety Blueprint to put teen safety first

OpenAI Blog · 3d ago