# Banks adopt location intelligence to detect fraud > Source: > Published: 2026-06-15 10:17:08.216842+00:00 # Banks adopt location intelligence to detect fraud The Citizen reports that banks in South Africa are adopting location intelligence to detect fraud by cross-referencing customer addresses, property records, and transaction locations with spatial datasets. The article quotes Marna Roos of AfriGIS: "Fraudsters will always follow the path of least resistance, and spatial reality is expensive to fake." Roos describes a persistent "spatial fingerprint" created from address entries and card-swipe locations and cites examples such as impossible street numbers or vacant plots tied to paper registrations. The piece frames location checks as a complement to document-based verification amid rising generative-AI risks that produce synthetic identities and forged paperwork. The reporting centers on Roos' remarks; no bank spokespeople are quoted in the article. ### What happened The Citizen reports that banks in South Africa are deploying **location intelligence** to strengthen fraud detection by cross-referencing customer-submitted addresses, property records, and transaction locations against spatial datasets. The article quotes Marna Roos of **AfriGIS**, saying, "Fraudsters will always follow the path of least resistance, and spatial reality is expensive to fake." Roos also describes a "spatial fingerprint" built from address entries and card-swipe locations that, over time, can corroborate whether a customer actually lives or operates where they claim. ### Technical Context Location intelligence typically combines geocoding, cadastral and deed registries, and transaction geolocation to create multi-source validation signals. Industry implementations often fuse point-in-polygon checks, address-parsing heuristics, and temporal consistency checks to surface anomalies such as non-existent street numbers or business registrations tied to vacant land. These methods trade on datasets that are difficult for attackers to fabricate at scale compared with static documents. ### Context and significance For practitioners, spatial signals act as a complementary modality to document and identity verification, particularly as generative AI enables higher-quality forged paperwork and synthetic identities. Industry observers note that data quality factors - geocoding accuracy, cadastral coverage, and dataset freshness - drive both detection performance and false-positive risk. ### What to watch For practitioners and vendors: integration of spatial checks into transaction pipelines, approaches to link mobility traces while preserving privacy, investments in authoritative cadastral feeds, and operational measures to tune for locality-specific address formats and edge cases. ## Scoring Rationale A practitioner-relevant piece on spatial fraud detection in South African banking, drawing on a single primary article quoting AfriGIS. The use case - location intelligence as a KYC complement against generative-AI-enabled synthetic identities - is directly on-topic for fraud and ML practitioners but has limited geographic scope and relies largely on a vendor-adjacent source. Practice with real Banking data 90 SQL & Python problems · 15 industry datasets [Suspicious Online TransactionsEasy](/problems/sql/suspicious-online-transactions) [Delinquent Loans Over 30 DaysMedium](/problems/sql/delinquent-loans-over-30-days) [Credit Card Utilization Risk ReportHard](/problems/sql/credit-card-utilization-risk-report) 250 free problems · No credit card [See all Banking problems](/problems/datasets/banking)