ITmedia AI+
Original published:
Is it dangerous to work on other tasks while logged into net banking? Verifying data theft with 7 AI browsers including 'ChatGPT Atlas'
AI summary
Editorial policyResearchers at the University of Washington published a paper titled 'Agentic Browsers and the Same-Origin Policy' verifying the risk of confidential data theft in AI browsers.
Key points
- •AI browsers where AI agents perform web operations on behalf of users were the subject of verification.
- •The study investigated the risk of confidential information such as emails and net banking being stolen by malicious sites.
- •The research was conducted by researchers affiliated with the University of Washington.