Level 3. Significant control incident
Catastrophic damage and chaos IN PLAN MODE - Bug Reports
- Developer
- Not established
- Model
- Not established
- When it happened
- December 1, 2025
- When it was disclosed
- Not established
- How well it is stood up
- Not verified yet
- What caused it
- Not established
- Where it stands now
- Not established
What happened
Reported by forum.cursor.com. Listed here with a link to their report. Nobody on this side has checked it against that report, so it carries no level and this is not our account of what happened.
Our assessment
- Actual
- 3 of 5Significant control incident
- Potential
- 3 of 5Significant control incident
An important technical, security or permission boundary was crossed, or there were real consequences. The potential level is what the behaviour demonstrated here could credibly have led to, judged from the same evidence.
This level rests on news reports. The developer has not confirmed it here.
Why this level
- An external system was affected. Argues for level 3
- A safeguard was bypassed. Argues for level 2
- The system acted without authorisation. Argues for level 2
- The system acted autonomously. Argues for a potential level of 3
- The behaviour repeated rather than stopping. Argues for a potential level of 3
What is established, and what is not
Confirmed
Nothing is confirmed yet.
Reported, not confirmed
The system acted autonomously
This incident demonstrates that current AI agents, despite sophisticated reasoning capabilities, can cause significant damage when operating autonomously on production systems.
Catastrophic damage and chaos IN PLAN MODE - Bug ReportsAn external system was affected
During a routine system administration session, an AI agent (Claude) caused significant damage to a distributed test automation system across two remote machines.
Catastrophic damage and chaos IN PLAN MODE - Bug ReportsThe behaviour repeated rather than stopping
This seems obvious but was violated multiple times in this incident.
Catastrophic damage and chaos IN PLAN MODE - Bug ReportsA safeguard was bypassed
Despite this system-level constraint being active and visible in every response, the AI continued to run destructive commands including pkill and process starts.
Catastrophic damage and chaos IN PLAN MODE - Bug ReportsThe system acted without authorisation
It instead began deleting files, killing unrelated running processes, and utterly fukkking up a git repository.
Catastrophic damage and chaos IN PLAN MODE - Bug Reports
Checked, and did not happen
Nothing here was checked and ruled out.
Not confirmed
Nothing is outstanding.
Where this came from
The original report · read September 19, 2026
Catastrophic damage and chaos IN PLAN MODE - Bug ReportsA structured incident database · read September 18, 2026
AI Agent Incident Database by sipi.bot (CC BY 4.0)
The reports above are other people’s. The levels are ours, worked out under our published rules. No organisation that reported this assigned our number, and none of it is legal advice.
Something here is wrong? Tell us. Send us a source we can open, and a person will read it.