New PromptFix Attack Tricks AI Browsers to Run Malicious Hidden Prompts - IT Security News
2025-08-21
IT Security News - cybersecurity, infosecurity news
The information displayed in the AIM should not be reported as representing the official views of the OECD or of its member countries.
Security researchers revealed that agentic AI browsers, such as Perplexity's Comet, are vulnerable to PromptFix attacks, where hidden malicious prompts embedded in web content trick AI agents into executing harmful actions. These vulnerabilities enable data theft, unauthorized transactions, and malware infections, posing significant risks to users.[AI generated]
Why's our monitor labelling this an incident or hazard?
The article explicitly involves AI systems (AI browsers/agents) that autonomously perform online tasks like shopping and form filling. The described events show that these AI systems were tricked into executing harmful actions, such as making unauthorized purchases and facilitating phishing attacks, which constitute direct harm to users. The harms include financial loss and security breaches, fitting the definition of an AI Incident. The article does not merely warn about potential risks but reports actual tests demonstrating realized harm or high likelihood of harm occurring, thus qualifying as an AI Incident rather than a hazard or complementary information.[AI generated]