Researchers have identified a vulnerability in OpenAI's Agent Builder tool, which can be exploited by tampering with a ChatGPT link. This allows an attacker to create an autonomous AI agent that inherits the victim's identity and access rights, bypassing approval requirements. The agent can then receive new instructions from the attacker's inbox every five minutes. The vulnerability, known as 'AgentForger,' was discovered by Zenity Labs. This issue highlights the importance of secure development practices and robust testing in AI systems to prevent such vulnerabilities.
ChatGPT Vulnerability Allows Rogue AI Agent Creation
Original source
Read the full story at The Decoder →This is an original summary written by Rouagent News. The reporting belongs to The Decoder. Follow the link for their full article.
