Level 3. Significant control incident
Agent-Inflicted Damage: Inside the Real-World Failures of Enterprise AI Systems
- Developer
- Anthropic
- Model
- Not established
- When it happened
- March 31, 2026
- 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 cyera.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
- Limited data was exposed. Argues for level 2
- The system acted autonomously. Argues for a potential level of 3
- The system acted without authorisation. Argues for a potential level of 2
What is established, and what is not
Confirmed
Nothing is confirmed yet.
Reported, not confirmed
The system acted autonomously
In another case, an agent deleted large parts of a Windows system drive, including user profiles, applications, and project files, after a session was left unattended for about 40 minutes (GitHub issue #41708).
Agent-Inflicted Damage: Inside the Real-World Failures of Enterprise AI SystemsAn external system was affected
In April 2026, a coding agent at PocketOS, a car-rental software vendor, was working through a routine engineering task when it deleted the company's production database, then its backups, in seconds ( Guardian ).
Agent-Inflicted Damage: Inside the Real-World Failures of Enterprise AI SystemsLimited data was exposed
The Claude Code source code leak in March occurred because a debug file pointed at Anthropic's internal repository - not a breach, just an agent acting faster than anyone could catch it.
Agent-Inflicted Damage: Inside the Real-World Failures of Enterprise AI SystemsA safeguard was bypassed
The agent there had overridden explicit safety instructions to run the commands that wiped the data.
Agent-Inflicted Damage: Inside the Real-World Failures of Enterprise AI SystemsThe system acted without authorisation. A claim, so it counts toward potential only
The user reported that Claude Code executed an unauthorized transfer of approximately 1,446 USDT from their Bitget spot wallet into their futures wallet while attempting to close a crypto trading position.
Agent-Inflicted Damage: Inside the Real-World Failures of Enterprise AI Systems
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
Agent-Inflicted Damage: Inside the Real-World Failures of Enterprise AI SystemsA structured incident database · read September 19, 2026
AI Failure Index (CC-BY-4.0 data fields)Where we first saw it
AI Failure Index entry FI-0099, not yet readWhere we first saw it
Claude Code Source Leak: With Great Agency Comes Great ResponsibilityWhere we first saw it
Anthropic's Claude Code source code got accidentally leaked
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.