Classified examples · Credit, fraud and insurance · Po polsku
AI flags suspicious payments and fraud
Minimal riskDeployer
A payment or e-commerce system uses AI to flag possibly fraudulent transactions for review or to block them.
How we classify it
- Annex III point 5(b) makes credit scoring high-risk but expressly excludes AI systems used to detect financial fraud.
- Fraud detection is not otherwise listed and is not prohibited.
- Result: not high-risk under the AI Act; GDPR and payment rules still apply.
What applies now
- GDPR: legal basis, information and a way to contest blocked transactions
- Human review of blocks that seriously affect customers
- Register entry and owner
What would change the answer
- If the same model is used to decide on credit or pricing for individuals, that use is high-risk
Evidence to keep
- Register entry
- Classification record noting the fraud exclusion
In the AI Compliance Kit: 03AI System Inventory · 04Risk Classification Procedure.
Dates and sources
- AI literacy duty, Article 4: 2 February 2025
Legal basis: AI Act Annex III point 5(b); GDPR.
Similar examples
- AI scores consumers' creditworthiness High-risk (Annex III)
- AI prices life or health insurance for individuals High-risk (Annex III)