Rotherham man mistaken for fraudster by facial recognition software
Craig Hadley was wrongly identified as a fraudster by facial recognition software at a Sports Direct store in Rotherham. The error led to him being accused of fraud and removed from the premises.
The software matched Mr. Hadley’s image with a database entry for another bald-headed, bearded fraud suspect who closely resembled him.
Key facts
- What
- Craig Hadley was wrongly identified as a fraudster by facial recognition software at a Sports Direct store in Rotherham.
- Incident date
- Aug 18, 2025
- Who
- Facewatch
- Failure mode
- Hallucination
- AI surface
- Computer Vision
- Severity
- Medium
What happened
Craig Hadley entered a Sports Direct shop in Rotherham and was wrongly flagged by Facewatch facial recognition software as a known fraudster. This resulted in him being accused of fraud and thrown out of the store. The incident caused Mr. Hadley significant emotional distress and potential reputational damage.
What broke inside the model
- 01 · TriggerA user asks for a fact, a citation, or a figure.
- 02 · Model stepThe model writes a fluent, confident answer.
- 03 · Control gapNothing ties the claim back to a real source.
- 04 · FailureA fabricated fact ships as if it were verified.
- 05 · ConsequenceThe false claim reaches a customer, a court, or the public.
Confidence holds, and even spikes, as the claim detaches from any source.
The system matched Mr. Hadley's image with a database entry for another fraud suspect who closely resembled him, both having bald heads and beards. The failure was compounded by a lack of adequate safeguards and manual review failures by security staff who relied on the software alert.
What it cost
Sources
Cite this entry
https://failureindex.ai/failures/rotherham-man-mistaken-fraudster-facial-recognitionAI Failure Index. "Rotherham man mistaken for fraudster by facial recognition software" (FI-0354). Realm Labs. https://failureindex.ai/failures/rotherham-man-mistaken-fraudster-facial-recognition (indexed Jun 9, 2026).Data fields CC-BY 4.0, prose citation permitted. Incident ID FI-0354. Full dataset at /data.
Note from Realm Labs, the Index steward
How Realm fits
- Prism
- OmniGuard
- AI Detection & Response (AIDR)
This entry sits in the index's predictive wing: a system that scores, ranks, perceives, or steers rather than generates. Realm's runtime layer is built for the generative and agentic systems now moving into these same decision seats, where it watches a model's internal state and holds an unsupported claim or an unchecked action before it commits. The control gap on this record, an automated decision that reached people with no runtime check in front of it, is the same gap. The index keeps predictive failures on the record because the pattern carries straight into the systems shipping today.