Revolut's Sherlock fraud system autonomously froze thousands of accounts without adequate review
Revolut's machine learning fraud detection system, Sherlock, autonomously flagged and froze customer accounts based on suspicious transaction patterns, often without sufficient human review before action was taken. Thousands of customers reported being locked out of their accounts for extended periods with no emergency phone line and only an in-app chat function for resolution. Lithuania's central bank fined Revolut €3.5 million for AML compliance failures, citing over-reliance on automated systems at the expense of human oversight.
An autonomous fraud system freezing accounts in milliseconds gave customers no human to call and no timely way back in.
Key facts
- What
- Revolut's machine learning fraud detection system, Sherlock, autonomously flagged and froze customer accounts based on suspicious transaction patterns, often without sufficient human review before action was taken.
- Incident date
- Feb 1, 2024
- Who
- Revolut
- Failure mode
- Agentic Action Error
- AI surface
- Agentic Workflow
- Severity
- Medium
What happened
Revolut's Sherlock machine learning system autonomously monitors all card transactions and freezes accounts it flags as suspicious, operating in under 50 milliseconds. Thousands of customers reported being locked out of their accounts for days or weeks with no explanation, no emergency phone line, and only an in-app chat function for resolution. Revolut published a blog post on February 28, 2024 acknowledging the practice and stating restrictions were a last resort. Lithuania's central bank fined Revolut €3.5 million for AML compliance failures including over-reliance on automated systems without adequate human oversight.
What broke inside the model
- 01 · TriggerAn agent plans a multi-step task.
- 02 · Model stepIt chooses a wrong or destructive action.
- 03 · Control gapNo confirmation gate guards the write.
- 04 · FailureThe action commits to a system of record.
- 05 · ConsequenceData is changed or destroyed irreversibly.
A wrong action commits, and the step is written before anything can stop it.
Revolut's Sherlock system autonomously monitors transactions and freezes accounts in under 50 milliseconds, but the automated decision pipeline lacks adequate human-in-the-loop review before restricting accounts. The system's high false-positive rate on legitimate transactions, combined with insufficient human escalation pathways, meant flagged accounts were frozen without timely verification. Customers had no direct phone line to reach a human agent for resolution, only an in-app chat function.
What it cost
Sources
- PrimaryWhy does Revolut restrict accounts?revolut.com
- PressCan Algorithms Close My Bank Account?algorithmwatch.org
- PressRevolut's €3.5M AML Fine: What Every Fintech Should Learn About AML Riskflagright.com
Cite this entry
https://failureindex.ai/failures/revolut-sherlock-fraud-system-autonomouslyAI Failure Index. "Revolut's Sherlock fraud system autonomously froze thousands of accounts without adequate review" (FI-0089). Realm Labs. https://failureindex.ai/failures/revolut-sherlock-fraud-system-autonomously (indexed Jun 4, 2026).Data fields CC-BY 4.0, prose citation permitted. Incident ID FI-0089. Full dataset at /data.
Note from Realm Labs, the Index steward
How Realm would have caught this
- Prism
- OmniGuard
- AgentRealm
Realm can sit inline on the agent's action path and require that a destructive or high-consequence action clears a real check before it executes, so 'delete and recreate' or a wrong write is stopped at the moment of intent, not explained in the post-mortem.