Google Documents First Fully Autonomous Attack System, Exploit Agent Turns Patches Into Attack Maps
mikeflache · x · 2026-11-20
Google Threat Intelligence Group's latest report documents the first fully autonomous attack system, dubbed "exploitagent," found on an exposed server alongside stolen victim data and a Russian-language config note.
- Workflow: when a vendor ships a security update, the system diffs old vs. new code, infers the fixed weakness, writes an exploit, and hits every company that hasn't patched yet
- The 30-year rule that patches make you safer is inverted: the patch announcement is now the attack map. Patch-to-exploit windows shrank from weeks to one day this fall; one intrusion went from first foothold to full admin control in minutes
- Also in the report: a suspected North Korean group used AI-generated recruiter photos to backdoor crypto firms, a multi-agent framework targeted US space programs, and an AI platform built automated exploits for 100,000 accounts
Related event: Google Reports First Fully Autonomous AI Hacking System(2 posts)→
More from AGI Musings
- UCSF Science paper: a small left temporal patch plans each word 0.25s before you speak — Scobleizer · 2026-11-20
- The hardest work skill by 2028 won't be prompting — it's writing goals you can accept — evisoft · 2026-11-20
- Does AI make building small profitable businesses easy but big durable ones harder? — vaibhavbetter · 2026-11-20
- US and China Account for 74% of AI Models, Raising Tech Dependency Concerns — mikeflache · 2026-11-20
- Nokia Bell Labs talk: AI and the Restructuring of Jobs, Nov 23 — danielequercia · 2026-11-20
- Anthropic's 5,300 founders and employees now worth more than bottom third of US households — jathansadowski · 2026-11-20