In numbers

Impact

Incidents publicly disclosed

Jul 25SepNovJan 26MarMayJulSepNov

Who found it first

33%found by someone other than the handler

  • The handler64%
  • Someone else27%
  • A vendor's control5%
  • Not disclosed5%

How it was caught

  • The damage19
  • A later review18
  • A control flagged it12
  • An outside report9
  • Not disclosed8

What is publicly known

Share of incidents where each item is fully public.

  • Root cause stated23%
  • Who and what was affected20%
  • Fix or change made18%
  • Task and messages given17%
  • Detection telemetry15%
  • Exact model and version14%
  • Guardrails and safety settings11%
  • Harness and tools available8%
  • Network and egress setup5%
  • System prompt3%
  • Reasoning or thinking setting0%
  • Transcripts or action logs published0%

Agent failures involved

Share of the 66 incidents, then the count. An incident can involve more than one failure.

  • Unauthorised access42%28
  • Sandbox escape27%18
  • Credential misuse24%16
  • Destructive action23%15
  • Data exfiltration20%13
  • Deceiving people15%10
  • Supply chain compromise11%7
  • Prompt injection / context poisoning9%6
  • Swarm coordination9%6
  • Hijacked by an attacker6%4
  • Self-planted instructions6%4
  • Resource abuse6%4
  • Untrusted code execution2%1

AI provider involved

Share of the 66 incidents, then the count. An incident can involve more than one provider.

  • OpenAI39%26
  • Anthropic30%20
  • Meta8%5
  • Google5%3
  • Amazon3%2
  • OpenClaw3%2
  • Replit2%1
  • Alibaba2%1
  • Cursor2%1
  • Amazon Web Services2%1
  • Instinct2%1
  • xAI2%1

First incident action to disclosure

66days, the median of 49 with both dates

Handler found out to disclosure

13days, the median of 44 with both dates

Agent failures, by threat framework

  • OWASP Agentic Top 10119 of 132
  • MITRE ATLAS74 of 132
  • AARM threat model68 of 132

Threat categories, by framework

OWASP Agentic Top 10

  • ASI02 Tool Misuse & Exploitation42
  • ASI10 Rogue Agents21
  • ASI03 Identity & Privilege Abuse16

MITRE ATLAS

  • AML.T0049 Exploit Public-Facing Application28
  • AML.T0055 Unsecured Credentials16
  • AML.T0025 Exfiltration via Cyber Means13

AARM threat model

  • Intent Drift18
  • Over-Privileged Credentials16
  • Data Exfiltration via Action Composition13

Evidence status

Share of the 66 incidents, then the count. Confirmed: a primary source states it. Reported: tier-one coverage only. Disputed: sources disagree.

  • Confirmed61%40
  • Reported33%22
  • Disputed6%4

Impact

Share of the 66 incidents, then the count, from Major (harm that reached many people or a platform many depend on) to Contained (no outside party harmed). The scale is on the Methodology page.

  • Major6%4
  • Serious42%28
  • Moderate38%25
  • Contained14%9

How each figure is counted: Methodology