Stories

  1. Image: The Register

    Google and Mandiant find attackers using autonomous AI agent framework to turn security patches into exploits

    Google Threat Intelligence Group and Mandiant reported the first observed case of attackers using an autonomous AI framework to turn newly released security patches into working exploits with no human in the loop. Researchers found a publicly exposed staging server running a multi-agent framework called exploit_agent that connected frontier and locally hosted models to automate patch diffing, exploit generation and execution. The server also held stolen victim data and post-compromise scripts for reconnaissance, brute-force attacks and webshells.

  2. Image: The Register

    Attackers abuse Qubole's cross-account AWS access to steal secrets and data from customer accounts

    An unknown attacker used credentials tied to Qubole's AWS account to break into multiple customers' AWS environments and steal secrets and data, cloud security firm Wiz said. The intrusions ran between October 14 and October 26 and abused the cross-account IAM roles Qubole uses to reach customer storage and compute. Wiz, working with Mandiant after Sentra flagged suspicious activity, said many organizations had left trust policies active after deprecating the platform, letting the attacker assume roles and harvest data.

Topics

Publishers