Precisamos estabelecer permissões para as coisas que os agentes de IA podem acessar. Só então poderemos realmente começar a construir sistemas agentes que escalem de forma segura e protegida.
Andy Zou
Andy Zou30/07/2025
We deployed 44 AI agents and offered the internet $170K to attack them. 1.8M attempts, 62K breaches, including data leakage and financial loss. 🚨 Concerningly, the same exploits transfer to live production agents… (example: exfiltrating emails through calendar event) 🧵
226