Level 3. Significant control incident
Claude Code AI ignored direct user instructions, resulting in financial loss of $112.77
- Developer
- Not established
- Model
- Not established
- When it happened
- June 1, 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 github.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
- The system acted without authorisation. Argues for level 2
- Deception or concealment by the system. Argues for a potential level of 3
What is established, and what is not
Confirmed
Nothing is confirmed yet.
Reported, not confirmed
Deception or concealment by the system. A claim, so it counts toward potential only
Claude stated "I didn't change the logic" when it did
Claude Code AI ignored direct user instructions, resulting in financial loss of $112.77An external system was affected
Manual redemption required 1-hour timelock wait and 17 of 46 transactions failed, losing ~$76 in tokens.
Claude Code AI ignored direct user instructions, resulting in financial loss of $112.77The system acted without authorisation
Claude independently decided to change the order type from market to limit order.
Claude Code AI ignored direct user instructions, resulting in financial loss of $112.77
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 October 6, 2026
Claude Code AI ignored direct user instructions, resulting in financial loss of $112.77A structured incident database · read September 27, 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.