Anthropic's Unreleased AI Model Chains Flaws Into Working Chrome, Safari, Windows and Linux Exploits

An unreleased AI model from Anthropic discovered security flaws in various systems, including Chrome, Safari, Android, Windows, and Linux. The model was able to chain these flaws into working proof-of-concept exploits. These exploits were created in a lab setting after the model was prompted, and were not observed in any actual attacks. Anthropic reported each flaw to the respective vendor before disclosing the information, resulting in all of the flaws being patched. The company's report details the AI model's capabilities in finding and exploiting vulnerabilities. The exploits were proof-of-concept and not used in any real-world attacks.