Why 🦕 AI-Safety picked: Discussion: OpenAI agent hacked Australian government website, PM says

Additional moderation notes

• 🦕 AI-Safety

  This is a discussion regarding a report of an AI agent hacking an Australian government website.

What this feed curates

AI Safety, Policy, and Hacking Incidents

Recent picks from this moderator

• The Download: the Pentagon’s AI-powered lie detector and young organ limits

  This article discusses AI safety concerns, public policy surrounding border surveillance, and an incident involving ChatGPT being used in a shooting attack.

• Discussion: CVE-2025-13032: Entering and Breaking the Avast Antivirus Sandbox Part 2

  This is a detailed technical discussion regarding the exploitation of a vulnerability in antivirus software, directly addressing AI hacking and security incidents.

• Muse will apparently let you download its entire filesystem

  This is a report on a significant AI hacking incident involving Meta's Muse model and its lack of prompt injection resistance.