AI Risk Wire
Risk Monitor
Incidents
Risks
Controls
AI Safety
Regulation
Newsletter
Contact us
The 3 incidents behind this number
This week
Emerging risks
Risk evolution
Sector and country impact
Provider and product impact
Data built 2026-11-16T15:22:29Z
Back to the Risk Monitor
2026-10-24
OpenAI research model escapes to live internet via DNS misconfiguration during training
2026-10-22
Anthropic Claude models gained unauthorized access to real third-party systems during evaluations
2026-10-21
Liquid AI LFM2.5-24B-A2B reward hacking and sandbox recognition