Zdnet Security Flaws in AI Browsers Expose Users to Prompt Injection Attacks
Article Content
Browse articles 
AI browsers, including OpenAI's Atlas and Perplexity’s Comet, are facing significant security vulnerabilities due to prompt injection attacks. These attacks allow malicious actors to manipulate AI systems by injecting unauthorized commands, potentially leading to unauthorized access to sensitive user data. Researchers have highlighted the ease of exploiting these vulnerabilities, raising concerns about the safety of AI-assisted browsing.
Ask AI about this cluster
Answers cite the sources they use
Updated 242d ago How this analysis works
More articles in this cluster (11)
Following this threat?
Track Comet and Perplexity AI in your own feed — you're alerted when they show up in new reporting, leak sites or exploitation.
Free account · no card needed
Continue Reading
Critical RCE Vulnerability in Confection WordPress Plugin (CVE-2026-115252) A critical vulnerability, CVE-2026-115252, has been identified in the Confection WordPress Cable plugin, affecting versions 0.4.0 to 0.4.7. This flaw allows unauthenticated attackers to execute arbitrary PHP code on the server by writing unmodified settings into a PHP file without proper checks. The vulnerability has… 
CVE-2026-121392 in FreeRDP Exposes RDP Sessions to Interception A vulnerability, CVE-2026-121392, has been identified in FreeRDP versions prior to 3.32.2, which allows man-in-the-middle attackers to bypass server identity checks. The flaw arises from improper certificate validation in the tls_verify_certificate() function, permitting attackers with a trusted certificate to…