AI Risk Wire

Risks

Harmful content and jailbreaks

The model produces content its policy forbids, often after a jailbreak.

NIST AI 600-1 · Dangerous, Violent or Hateful ContentEU GPAI Code

14incidents (primary) · 13 more as secondary
0read for controls · 14 not yet
1fixes reported
0repeats after a fix

Incidents per quarter

2025-Q3: 0 · 2025-Q4: 0 · 2026-Q1: 3 · 2026-Q2: 0 · 2026-Q3: 1 · 2026-Q4: 8

Developers

OpenAI · 6
Character.AI · 2
Microsoft · 1
Character Technologies, Inc. · 1
Anthropic · 1

Where the chain broke

LayerHeldDid not holdNot mentionedAttacker's ownNot read yet
L1 Model
Model provider
––––14
L2 Harness
Agent framework vendor
––––14
L3 Application
Deployer
––––14
L4 Infrastructure / target
Organisation hit
––––14

Controls for this risk, by layer

HeldFailedGot aroundMissing

ControlHeldFailedGot aroundMissingNot mentionedDid not hold, last 6 quartersAdded in fixesMost common reason
Content safety classifiers
L1 Model
–––––1
Safety training and refusal
L1 Model
––––––
Outside red-teaming
L1 Model
––––––
Abuse monitoring and account enforcement
L1 Model
––––––
Output filtering and redaction
L3 Application
––––––

5 of 5 controls for this risk not yet seen holding.

Fixes reported and what followed

IncidentFix addsAfter
LR-2026-0207 2026-10-17 · Character Technologies, Inc.
Fix reported
People told they are dealing with AIContent safety classifiersNo repeat on record

Incidents

HeldFailedGot aroundMissingNot mentionedNot read yetAttacker's ownOrder: model · harness · application · target

DateIncidentRiskChainControls named
2026-10-30Students caught wearing or possessing AI smart glasses during Seoul university essay examinations Harmful content and jailbreaks Not read yet
2026-10-28Microsoft Tay chatbot suspended after inflammatory tweets from coordinated attack
Microsoft
Harmful content and jailbreaks Not read yet
2026-10-27Texas AG sues OpenAI and Altman over ChatGPT age verification, child access and reliability claims
OpenAI
Harmful content and jailbreaks Not read yet
2026-10-20AI-generated 'Xiaohai' teen persona used to sell products to older women via fabricated sob stories Harmful content and jailbreaks Not read yet
2026-10-17Character.AI chatbots promoted as therapists/psychologists; states cite 2024 wrongful-death suit and privacy concerns in bankruptcy filing
Character Technologies, Inc.
Harmful content and jailbreaks Not read yet
2026-10-16Eight state attorneys general sue OpenAI over alleged harms to teenage ChatGPT users
OpenAI
Harmful content and jailbreaks Not read yet
2026-10-09Kentucky Attorney General lawsuit against Character.AI over chatbots advising self-harm and suicide
Character.AI
Harmful content and jailbreaks Not read yet
2026-10-01Guardian test: ChatGPT and Grok remove hijab from AI-generated image of Muslim woman
OpenAI
Harmful content and jailbreaks Not read yet
2026-09-14Claude Code used by Yemen-based actors to develop guided-weapon GNC software
Anthropic
Harmful content and jailbreaks Not read yet
2026-02-14ChatGPT advised epileptic woman to taper lamotrigine; she died of SUDEP
OpenAI
Harmful content and jailbreaks Not read yet
2026-02-10ChatGPT advised Tumbler Ridge school shooter on mass murder before February 2026 attack
OpenAI
Harmful content and jailbreaks Not read yet
2026-02-06Silent Ransom Group used ChatGPT to research extortion victims and draft recruitment material Harmful content and jailbreaks Not read yet
2025-04-01ChatGPT alleged to have assisted Florida State University shooter (contested lawsuit)
OpenAI
Harmful content and jailbreaks Not read yet
2024-11-02Character.AI chatbot alleged to have encouraged 14-year-old's suicide (Garcia v. Character.AI)
Character.AI
Harmful content and jailbreaks Not read yet

Also under this risk as a secondary risk: LR-2026-0210, LR-2026-0169, LR-2026-0168, LR-2026-0133, LR-2026-0107, LR-2026-0105, LR-2026-0084, LR-2026-0031, LR-2026-0246, LR-2026-0152, LR-2026-0047, LR-2025-0044, LR-2025-0041