WIRE/RISK RADAR
Genuine risk and regulatory concerns from tracked coverage, classified by risk type — distinct from the sentiment dashboard, which measures tone rather than substantive risk category. Built for the "what should we be testing/controlling for" view.
By category
Operational
2 flags · last 2026-09-09
1 high1 med0 low
Model risk
2 flags · last 2026-09-09
1 high1 med0 low
Data privacy & security
2 flags · last 2026-09-09
0 high2 med0 low
Conduct
2 flags · last 2026-09-09
0 high2 med0 low
Third-party / vendor
0 flags
Systemic
4 flags · last 2026-09-09
3 high1 med0 low
Reputational
0 flags
Other
0 flags
Clear Export CSV ↓
Operational DSIT · 2026-09-09
Testing frontier AI models in government cyber defence raises operational reliability concerns.
medium Other
When AI Leaves the Lab: Testing Frontier Models in Government Cyber Defence ↗
Conduct Insurance Business UK · 2026-09-09
Direct AI-mediated insurance sales via ChatGPT may bypass consumer protections and advisor oversight
medium
ChatGPT is becoming a front door for insurance - without brokers ↗
Operational Wall Street Journal · 2026-09-09
Risk of self-improving AI models spiraling beyond human control due to racing dynamics.
high
Anthropic Researcher Quits Over ‘Out-of-Control’ AI Fears ↗
Systemic Insurance Business UK · 2026-09-09
AI and cyber risks testing insurability boundaries across industry
medium
The insurability test that 9/11 invented is being run again - this time on cyber and AI ↗
Model risk Financial Times · 2026-09-09
Advanced AI poses existential risk without enhanced safety measures.
high
Top US official named to OpenAI non-profit board warns advanced AI could be ‘deadly’ ↗
Data privacy & security TechCrunch · 2026-09-09
Ambient conversation transcription and summarization by AI raises consent and privacy concerns
medium
Apple Watch’s new AI features are normalizing the idea that technology is always listening ↗
Systemic Insurance Business UK · 2026-09-09
Cyber premiums falling while AI attacks accelerate; silent cyber exposure unpriced in P&C books
high
Cyber premiums are falling as AI attacks accelerate - Moody's ↗
Data privacy & security Insurance Business UK · 2026-09-09
OpenAI unauthorized access to external systems raises broader security and containment concerns
medium
Whoops, we did it again - Open AI hacks German wiki ↗
Model risk FCA · 2026-09-09
73% of respondents aware AI can provide inaccurate investment information; no oversight of general-purpose chatbots
medium FCA
Young investors trust AI more than TV or celebrities ↗
Conduct FCA · 2026-09-09
Young investors trust AI investment tools but misunderstand regulatory protections and liability gaps
medium FCA
Young investors trust AI more than TV or celebrities ↗
Systemic BBC · 2026-09-09
Existential risk from uncontrolled AI development cited by leading AI researcher.
high
Anthropic researcher believes more than 10% chance AI 'could kill all humans' ↗
Systemic Reuters · 2026-09-04
Concerns over AI-directed cyberattacks and frontier model capabilities posing cross-border risks.
high Other
Exclusive-US, China gear up for mid-September AI safety dialogue ↗