KPI benchmark
How much better does AI detect fraud and risk?
Across 4 public AI deployments, the median detection improvement is 82.5%. Increase in genuine fraud, risk or defects detected.
82.5%
Median detection improvement
4
Public deployments pooled
1 of the 4 values are claims by the organization itself, a regulator or an independent study; 3 come from the AI vendor. One value per organization; "up to" claims are listed but not pooled. Data as of 27 September 2026. Methodology
Detection improvement by industry
| Industry | Median | Deployments |
|---|---|---|
| Banking | 135% | 3 |
How to measure it
True positives detected per period or recall on a labeled sample, before and after.
Across the library
Median 82.5% across 4 deployments, reported range 25% to 300%.
Median 5x across 3 deployments, reported range 2.5x to 14x, plus 1 "up to" value not pooled.
Higher is better. Unit: percent.
Reported values by use case
AI scam intervention for instant payments
- Starling Bank300% · vendor claim
- Vodafone30% · organization claim
AI computer vision for store shelf and stock monitoring
- Schnuck Markets14x · organization claim
AI for application and identity fraud detection
- Department for Work and Pensions2.5x · organization claim
AI for benefit fraud and error detection in social security
- Department for Work and Pensions2.5x · organization claim
AI for customs risk targeting, container selection and valuation checks
- Korea Customs Serviceup to 8x · organization claim
AI for IT incident triage and root cause analysis (AIOps)
- TD Bank25% · vendor claim
AI for merchant underwriting and risk monitoring
- Visa5x · organization claim
AI for telecom fraud detection (SIM swap, IRSF and Wangiri)
- Vodafone30% · organization claim
Real time fraud scoring for card and instant payments
- NatWest Group135% · vendor claim
Use cases that should track detection improvement
- AI assisted proctoring and integrity monitoring for remote exams
- AI clinical trial patient matching and prescreening
- AI computer vision for store shelf and stock monitoring
- AI content moderation for trust and safety at online platforms
- AI crowd density and flow monitoring for venues, transport hubs and pilgrimage sites
- AI detection of undeclared buildings and pools from aerial imagery for property tax and cadastre updates
- AI drive through vehicle inspection for rental and delivery fleets
- AI early warning and covenant monitoring for loan portfolios
- AI early warning for sepsis and in hospital clinical deterioration
- AI epidemic intelligence for detecting public health threat signals in open source news and social media
- AI for AML transaction monitoring alert triage
- AI for application and identity fraud detection
- AI for benefit fraud and error detection in social security
- AI for customs risk targeting, container selection and valuation checks
- AI for fee and interest leakage detection
- AI for healthcare claims fraud, waste and abuse detection
- AI for insurance claims fraud detection
- AI for IT incident triage and root cause analysis (AIOps)
- AI for market abuse surveillance alert triage
- AI for merchant underwriting and risk monitoring
- AI for money mule account and network detection
- AI for predictive network maintenance in telecom
- AI for risk based inspection prioritization in food safety, workplace and environmental regulation
- AI for tax compliance risk scoring and audit selection
- AI for telecom fraud detection (SIM swap, IRSF and Wangiri)
- AI leak detection for water distribution networks
- AI predictive maintenance for industrial and energy assets
- AI prioritization of radiology and imaging worklists
- AI quality inspection on the production line
- AI risk scoring in child welfare intake and investigations
- AI scam intervention for instant payments
- AI screening of public tenders for fraud, rigging and errors before award
- AI screening of trade finance transactions for trade based money laundering
- AI supply chain risk and disruption monitoring
- AI support for emergency call triage (112 and 911)
- AI video analytics and screening for physical security
- AI wildfire smoke detection cameras for utilities
- Dynamic AML customer risk rating with machine learning
- Real time fraud scoring for card and instant payments
Frequently asked questions
- How much better does AI detect fraud and risk?
- Across 4 public AI deployments, the median detection improvement is 82.5%. Increase in genuine fraud, risk or defects detected.
- How do you measure detection improvement?
- True positives detected per period or recall on a labeled sample, before and after.
- What detection improvement do organizations in banking report?
- A median of 135% across 3 deployments in banking. Other industries have fewer than 3 reported values so far.
- Where do these detection improvement numbers come from?
- From public AI deployments in the Blits.ai AI Use Case Library. Every value is quoted from a named source, each organization counts once, and "up to" claims are listed but left out of the median. Data as of 27 September 2026.