AI Risk Wire

AI Risk Monitor

778 total incidents captured
192 new in the last 30 days
2 risks rising
15 controls breachedin 22 incidents

AI used to run a cyber attack: 4.0 → 9.7 a month

Reports a month, each counted once: ○ Feb–Jul ● Aug–Oct · rising

0246810 AI used to run a cyber attack4.0 → 9.7Hidden instructions hijack the agent3.7 → 6.3Physical harm4.3 → 4.7Harmful content and jailbreaks5.2 → 4.0Agent takes destructive or unapproved actions4.5 → 3.7Credential and privilege abuse1.2 → 3.7AI system attacked or breached3.3 → 3.7Manipulation and misinformation3.0 → 3.0

Incidents recorded: real-world and lab tests · Last 12 months

How we count

Incidents made public in Nov 2025 – Oct 2026: real-world and lab tests, by month. Lab tests: incidents whose source states an evaluation, training or research setting.

Real-worldLab tests
Nov 2025 24Dec 2025 301 31Jan 2026 24Feb 2026 50Mar 2026 47Apr 2026 471 48May 2026 34Jun 2026 432 45Jul 2026 606 66Aug 2026 4314 57Sep 2026 6710 77Oct 2026 16321 184

Incidents confirmed against preliminary · Last 12 months

How we count

Real-world incidents made public in Nov 2025 – Oct 2026, by evidence level: Confirmed or Preliminary, by fixed rules about their sources.

Preliminary 496Confirmed 136

632 of 632 incidents state it.

Built 16 Nov 2026. Periods end with the last full week or month; the week or month in progress is not counted. Bold: the five largest rows in each chart, ties included; controls by incidents where the control broke. Linked numbers open the incidents or model versions they count.

Reported incidents, not rates · 75 sources polled hourly since Sep 2026; earlier incidents come from a one-off backfill · English only · stories told twice are not yet merged · our own severity grades are not counted · built 2026-11-16T15:22:29Z

Linked numbers open the incidents they count. Updated 2026-11-16T15:22:29Z