{
  "id": 11141,
  "url": "https://thenewstack.io/gpt-red-prompt-injection-testing",
  "title": "OpenAI’s GPT-Red automates prompt injection testing to harden AI agents",
  "summary": "Now that AI agents are performing real tasks rather than just generating text, the old methods of manual security testing The post OpenAI’s GPT-Red automates prompt injection testing to harden AI agents appeared first on The New Stack .",
  "authors": "Amanda Caswell",
  "category": "news",
  "topics": "agents-autonomy",
  "orgs": "openai",
  "regions": null,
  "published_at": "2026-07-16T14:58:18.000Z",
  "fetched_at": "2026-07-17T05:10:53.887Z",
  "source_slug": "x-the-new-stack-ai",
  "source_name": "The New Stack AI",
  "source_homepage": "https://thenewstack.io/ai/",
  "ethics_ai_record_url": "https://ethics.ai/record/11141",
  "original_url": "https://thenewstack.io/gpt-red-prompt-injection-testing",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}