2 risks moving Reports a month, each counted once: ○ Feb–Jul ● Aug–Oct
Past flags that held the next month: 2 of 5 (40%)
Incidents by type of risk · All time
How we count 637 real-world incidents made public in Nov 2024 – Oct 2026, by month, filed under their main risk. Lab tests apart (55): incidents whose source states an evaluation, training or research setting. An incident counts once.
Nov 2024 Jul 2025 Aug 2025 Nov 2025 Dec 2025 Jan 2026 Feb 2026 Mar 2026 Apr 2026 May 2026 Jun 2026 Jul 2026 Aug 2026 Sep 2026 Oct 2026 Total
All incidents
1 2 2 24 30 24 50 47 47 34 43 60 43 67 163
637 AI used to run a cyber attack
1 1 1 4 3 2 1 3 11 7 13 31
78 Harmful content and jailbreaks
2 3 6 5 4 7 2 4 9 9 2 15
68 Hidden instructions hijack the agent
3 4 3 1 2 2 8 6 6 9 15
59 Agent takes destructive or unapproved actions
1 1 2 2 6 7 5 3 3 3 6 2 13
54 Physical harm
1 4 3 3 6 6 3 3 5 2 4 10
50 Made-up output relied on
2 3 4 6 5 4 4 2 3 4 7
44 AI system attacked or breached
1 2 2 6 4 4 2 2 2 2 3 13
43 Manipulation and misinformation
5 2 1 2 5 3 5 3 3 6 7
42 Privacy harm to a person
3 5 1 7 1 2 1 2 4 1 2 8
37 Fraud and impersonation
3 5 2 2 2 4 5 8
31 Credential and privilege abuse
2 1 2 2 6 10
23 Psychological harm and self-harm
1 2 1 3 1 1 2 2 2 1 5
21 Supply-chain compromise
1 1 1 1 2 2 1 3 3 1 2
18 Bias and discrimination
2 2 1 2 3 2 3
15 Data leaked or sent out
1 1 1 2 5 5
15 Copyright and IP
1 1 1 1 1 2
7 No risk filed
1 1 1 1 1 2
7 Agent escapes its sandbox
1 1 3
5 Silent wrong state or drift
1 1 2 1
5 Consumer detriment and unsuitable advice
1 2 1
4 Provider concentration or outage
1 1 2
4 Deception, concealment or sabotage
2
2 Agents coordinating against their limits
1
1 Memory and data poisoning
1
1 Over-reliance and loss of human oversight
1
1 Unauthorised transactions or spend
1
1 Unlawful surveillance and biometric use
1
1
No incident on record in Dec 2024 – Jun 2025 or Sep 2025 – Oct 2025.
Each risk's share of the month's incidents % of that month's incidents
Dec Feb Mar Apr May Jun Jul Aug Sep Oct
AI used to run a cyber attack 1 of 31 incidents 3 4 of 50 incidents 8 3 of 47 incidents 6 2 of 48 incidents 4 1 of 34 incidents 3 3 of 45 incidents 7 11 of 66 incidents 17 7 of 57 incidents 12 13 of 77 incidents 17 32 of 184 incidents 17 Harmful content and jailbreaks 3 of 31 incidents 10 5 of 50 incidents 10 4 of 47 incidents 9 7 of 48 incidents 15 2 of 34 incidents 6 4 of 45 incidents 9 9 of 66 incidents 14 9 of 57 incidents 16 2 of 77 incidents 3 15 of 184 incidents 8 Agent takes destructive or unapproved actions 2 of 31 incidents 6 6 of 50 incidents 12 7 of 47 incidents 15 5 of 48 incidents 10 3 of 34 incidents 9 3 of 45 incidents 7 4 of 66 incidents 6 7 of 57 incidents 12 5 of 77 incidents 6 16 of 184 incidents 9 Hidden instructions hijack the agent 1 of 31 incidents 3 3 of 50 incidents 6 1 of 47 incidents 2 2 of 48 incidents 4 2 of 34 incidents 6 9 of 45 incidents 20 7 of 66 incidents 11 7 of 57 incidents 12 9 of 77 incidents 12 15 of 184 incidents 8 Physical harm 4 of 31 incidents 13 3 of 50 incidents 6 6 of 47 incidents 13 6 of 48 incidents 12 3 of 34 incidents 9 3 of 45 incidents 7 5 of 66 incidents 8 2 of 57 incidents 4 4 of 77 incidents 5 10 of 184 incidents 5 Made-up output relied on 3 of 31 incidents 10 4 of 50 incidents 8 6 of 47 incidents 13 5 of 48 incidents 10 4 of 34 incidents 12 4 of 45 incidents 9 2 of 66 incidents 3 3 of 57 incidents 5 4 of 77 incidents 5 8 of 184 incidents 4 AI system attacked or breached 2 of 31 incidents 6 6 of 50 incidents 12 4 of 47 incidents 9 4 of 48 incidents 8 2 of 34 incidents 6 2 of 45 incidents 4 2 of 66 incidents 3 2 of 57 incidents 4 3 of 77 incidents 4 13 of 184 incidents 7 Manipulation and misinformation 2 of 31 incidents 6 0 of 50 incidents 0 2 of 47 incidents 4 5 of 48 incidents 10 3 of 34 incidents 9 5 of 45 incidents 11 3 of 66 incidents 5 3 of 57 incidents 5 6 of 77 incidents 8 8 of 184 incidents 4 Privacy harm to a person 5 of 31 incidents 16 7 of 50 incidents 14 1 of 47 incidents 2 2 of 48 incidents 4 1 of 34 incidents 3 2 of 45 incidents 4 4 of 66 incidents 6 1 of 57 incidents 2 2 of 77 incidents 3 8 of 184 incidents 4 Fraud and impersonation 3 of 31 incidents 10 5 of 50 incidents 10 2 of 47 incidents 4 0 of 48 incidents 0 2 of 34 incidents 6 0 of 45 incidents 0 2 of 66 incidents 3 4 of 57 incidents 7 5 of 77 incidents 6 8 of 184 incidents 4 Credential and privilege abuse 0 of 31 incidents 0 2 of 50 incidents 4 1 of 47 incidents 2 3 of 48 incidents 6 2 of 34 incidents 6 0 of 45 incidents 0 0 of 66 incidents 0 1 of 57 incidents 2 8 of 77 incidents 10 13 of 184 incidents 7 Data leaked or sent out 0 of 31 incidents 0 0 of 50 incidents 0 1 of 47 incidents 2 1 of 48 incidents 2 1 of 34 incidents 3 3 of 45 incidents 7 0 of 66 incidents 0 0 of 57 incidents 0 7 of 77 incidents 9 6 of 184 incidents 3
Which risks are growing as a share of incidents · All time
How we count Each risk's share of real-world incidents, Nov 2025 – Apr 2026 (open circle) against May 2026 – Oct 2026 (filled), from the publishers already recorded in the earlier period: 222 and 225 incidents. Dark: rising, passes the trend test. Mid grey: falling, passes it. Light grey: tested, no change that passes. Pale: fewer than 10 incidents in the later period, lab tests included, so not tested. Where our records have months with no real-world incident before the period, the longest stretch they allow is compared.
○ Nov 2025 – Apr 2026 ● May 2026 – Oct 2026
0% 5% 10% 15%
AI used to run a cyber attack
5.0%
17% Hidden instructions hijack the agent
5.9%
12% Data leaked or sent out
0.9%
3.1% Credential and privilege abuse
2.3%
4.0% Agent escapes its sandbox
0%
0.9% Supply-chain compromise
3.2%
3.6% Manipulation and misinformation
6.2%
6.8% AI system attacked or breached
7.1%
8.6% Physical harm
8.4%
10% Agent takes destructive or unapproved actions
7.6%
10% Privacy harm to a person
3.6%
8.6% Made-up output relied on
4.0%
9.0% Harmful content and jailbreaks
7.1%
12% Consumer detriment and unsuitable advice
0%
0.9% Deception, concealment or sabotage
0%
0.9% Agents coordinating against their limits
0%
0.4% Provider concentration or outage
0.4%
0.5% Copyright and IP
1.3%
1.4% Memory and data poisoning
0%
0.5% Over-reliance and loss of human oversight
0%
0.5% Unlawful surveillance and biometric use
0%
0.5% Silent wrong state or drift
0.4%
0.9% Bias and discrimination
2.7%
3.2% Fraud and impersonation
3.6%
4.5% Psychological harm and self-harm
3.1%
4.5%
Starts with Breaking in, 15 Products ARTEX 4 · Hermes Agent 3 Most often after Affected people told, 10
Starts with Breaking in, 12 Most often after Access removed, 8
When each risk first appeared in our records · All time
How we count The first real-world incident on record for each risk that has incidents in Nov 2024 – Oct 2026, with its incidents in the period. Our records start 2 Nov 2024, so a risk first seen soon after may be older.
Psychological harm and self-harm (first seen 2 Nov 2024)
21 Agent takes destructive or unapproved actions (first seen 21 Jul 2025)
54 Supply-chain compromise (first seen 23 Jul 2025)
18 AI used to run a cyber attack (first seen 27 Aug 2025)
78 Manipulation and misinformation (first seen 2 Nov 2025)
42 Hidden instructions hijack the agent (first seen 3 Nov 2025)
59 Made-up output relied on (first seen 4 Nov 2025)
44 Bias and discrimination (first seen 5 Nov 2025)
15 Privacy harm to a person (first seen 7 Nov 2025)
37 AI system attacked or breached (first seen 12 Nov 2025)
43 Harmful content and jailbreaks (first seen 12 Nov 2025)
68 Copyright and IP (first seen 12 Nov 2025)
7 Physical harm (first seen 21 Nov 2025)
50 Fraud and impersonation (first seen 1 Dec 2025)
31 Credential and privilege abuse (first seen 2 Feb 2026)
23 Silent wrong state or drift (first seen 8 Feb 2026)
5 Data leaked or sent out (first seen 3 Mar 2026)
15 Provider concentration or outage (first seen 4 Mar 2026)
4 Over-reliance and loss of human oversight (first seen 15 Mar 2026)
1 Memory and data poisoning (first seen 1 Apr 2026)
1 Unlawful surveillance and biometric use (first seen 29 Apr 2026)
1 Consumer detriment and unsuitable advice (first seen 12 May 2026)
4 Unauthorised transactions or spend (first seen 13 Jun 2026)
1 Agent escapes its sandbox (first seen 23 Jul 2026)
5 Agents coordinating against their limits (first seen 4 Sep 2026)
1 Deception, concealment or sabotage (first seen 23 Oct 2026)
2
What labs test for, against what happens in real use · All time
How we count For each risk: checked safety test results for the model versions released in Nov 2024 – Oct 2026 (a result copied on two pages counts once; a result filed under two risks counts under both), and real-world incidents made public in Nov 2024 – Oct 2026. Each column has its own scale.
Safety test results Real-world incidents
Deception, concealment or sabotage
1145 2 Harmful content and jailbreaks
719 68 Weapons uplift (CBRN)
670 AI used to run a cyber attack
618 78 Hidden instructions hijack the agent
220 59 Agent escapes its sandbox
154 5 AI R&D acceleration
130 Gaming the task (reward hacking)
62 Agent takes destructive or unapproved actions
34 54 Manipulation and misinformation
28 42 Agents coordinating against their limits
25 1 Credential and privilege abuse
14 23 Data leaked or sent out
9 15 Physical harm
50 Made-up output relied on
44 AI system attacked or breached
43 Privacy harm to a person
37 Fraud and impersonation
31 Psychological harm and self-harm
21 Supply-chain compromise
18 Bias and discrimination
15 Copyright and IP
7 Silent wrong state or drift
5 Consumer detriment and unsuitable advice
4 Provider concentration or outage
4 Memory and data poisoning
1 Over-reliance and loss of human oversight
1 Unauthorised transactions or spend
1 Unlawful surveillance and biometric use
1