> Researchers exploit OpenAI's Atlas by disguising prompts as URLs

[SOURCE] The Register Security [AUTHOR: Richard Speed] [DATE: 27/10/2025 13:54] [LANGUAGE: EN]

NeutralTrust shows how agentic browser can interpret bogus links as trusted user commands Researchers have found more attack vectors for OpenAI's new Atlas web browser – this time by disguising a potentially malicious prompt as an apparently harmless URL.…

[messages.read_original_source] →