KPI benchmark
How much does AI reduce fraud losses?
Across 5 public AI deployments, the median fraud loss reduction is 40%. Reduction in money lost to fraud or scams.
40%
Median fraud loss reduction
5
Public deployments pooled
4 of the 5 values are claims by the organization itself, a regulator or an independent study; 1 come from the AI vendor. One value per organization; "up to" claims are listed but not pooled. Data as of 30 September 2026. Methodology
Fraud loss reduction by industry
| Industry | Median | Deployments |
|---|---|---|
| Banking | 58% | 4 |
Fraud loss reduction by use case
| Use case | Median | Deployments |
|---|---|---|
| AI scam intervention for instant payments | 40% | 3 |
| Real time fraud scoring for card and instant payments | 30% | 3 |
How to measure it
Gross fraud losses per period, normalized for volume, before and after.
Across the library
Median 40% across 5 deployments, reported range 30% to 82%.
Higher is better. Unit: percent.
Reported values by use case
AI scam intervention for instant payments
- Commonwealth Bank of Australia76% · organization claim
- Česká spořitelna40% · organization claim
- Revolut30% · organization claim
Real time fraud scoring for card and instant payments
- Commonwealth Bank of Australia76% · organization claim
- Revolut30% · organization claim
- Stripeat least 30% · organization claim
AI agent for fraud alert confirmation with cardholders
- Commonwealth Bank of Australia76% · organization claim
- Revolut30% · organization claim
AI for money mule account and network detection
- bunq82% · vendor claim
Use cases that should track fraud loss reduction
- AI agent for fraud alert confirmation with cardholders
- AI agent for payment initiation within a customer mandate
- AI blocking of phishing and scam text messages in the operator network
- AI for application and identity fraud detection
- AI for benefit fraud and error detection in social security
- AI for insurance claims fraud detection
- AI for money mule account and network detection
- AI for telecom fraud detection (SIM swap, IRSF and Wangiri)
- AI order fraud screening for ecommerce checkout
- AI scam intervention for instant payments
- AI spam and scam call blocking for mobile and landline subscribers
- AI underwriting and verification for auto loans
- Real time fraud scoring for card and instant payments
Frequently asked questions
- How much does AI reduce fraud losses?
- Across 5 public AI deployments, the median fraud loss reduction is 40%. Reduction in money lost to fraud or scams.
- How do you measure fraud loss reduction?
- Gross fraud losses per period, normalized for volume, before and after.
- What fraud loss reduction do organizations in banking report?
- A median of 58% across 4 deployments in banking. Other industries have fewer than 3 reported values so far.
- Which AI use cases report the best fraud loss reduction?
- Among use cases with at least 3 reported values: AI scam intervention for instant payments (median 40%, 3 deployments); real time fraud scoring for card and instant payments (median 30%, 3 deployments).
- Where do these fraud loss reduction 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 30 September 2026.