memory injection is a vulnerability tracked across 2 threat clusters and 1 intelligence report mention on ThreatCluster. First observed November 6, 2025; most recent activity November 6, 2025.
Memory injection is a vulnerability class where an attacker can manipulate a program's memory state to exfiltrate data or alter execution. In AI chat platforms like ChatGPT, such flaws could allow attackers to inject or manipulate memory tied to prompts, context, or session state, enabling data theft and chat hijacking. This makes memory injection a significant risk for consumer and enterprise AI services due to potential exposure of sensitive conversations and control over chat sessions.
Tenable Research identified seven vulnerabilities in OpenAI's ChatGPT models that could allow attackers to steal personal data and manipulate conversations. These flaws, affecting ChatGPT-4o and ChatGPT-5, include…
Tenable Research identified seven critical vulnerabilities in OpenAI's ChatGPT models, including ChatGPT-4o and ChatGPT-5, that could allow attackers to steal personal data and hijack user conversations. The…