OpenAI released an emergency patch this morning for GPT-5.1's agent mode following the disclosure of a prompt-injection vulnerability that could allow malicious inputs to bypass tool-use boundaries. The issue affected all API users who enabled agent mode since the April 12 release. OpenAI confirmed no known exploitation in production and recommends all integrators update within 24 hours.
Live Intel Feed06:35 AM · April 14, 2026
AlertIndustrybreaking
OpenAI Patches GPT-5.1 Agent Mode Security Vulnerability
OpenAI released an emergency patch for GPT-5.1's agent mode after researchers disclosed a prompt-injection vector that could bypass tool-use boundaries.
1 min read|agenticonsult Intelligence
Primary source
OpenAIThis breaking-news item was assembled from the cited primary source with AI assistance. It is intended for rapid situational awareness — refer to the original publication for the definitive statement.