AI Browsers Under Fire: Hidden Web Prompts Can Hijack Your Agent and Connected Accounts
Security researchers are warning that artificial intelligence (AI)-powered browsers and agents from Perplexity, OpenAI, and Anthropic face escalating risks of covert prompt injection attacks and privacy breaches, potentially exposing user data through connected accounts and APIs. AI Browser Vulnerabilities Raise Security Concerns AI browsers and agents from Perplexity, OpenAI, and Anthropic are redefining how users [...]