Bank of Ireland cuts customer payment fraud losses 25% with Featurespace

Payment fraud happens when fraudsters deceive people into sending money or gain access to accounts to make transactions without the customer's permission. Bank of Ireland's Head of Fraud Protection described protecting customers against fraud as one of the bank's top priorities and said staying ahead of fraudsters requires constant innovation. The bank's new approach aims to spot suspicious activity before a payment leaves the bank while reducing unnecessary alerts that can delay payments.

Bank of Ireland implemented new payment fraud measures in 2024 built on AI-led behavior pattern recognition technology, delivered in partnership with Featurespace, a Visa Solution. The Featurespace platform uses AI and machine learning to understand how accounts are typically used, which helps the bank identify potentially suspicious behavior earlier. It is designed to identify both existing and new forms of fraud.

Since 2024, the technology has achieved a 30% reduction in attempted payment fraud and a 25% reduction in customer losses. Case handling time was halved, false positives dropped 87.1% and alert volumes fell 85%, and 90% of customers confirm genuine transactions without contacting the bank. The figures come from Bank of Ireland's internal data as used in its Celent Model Risk Manager 2026 application, and the results were recognized by Celent's 2026 Model Risk Manager Award.

• Models that learn how accounts are typically used can flag suspicious payments before they leave the bank.
• Fewer false positives help customers as well as fraud teams, because fewer genuine payments need to be confirmed.
• The bank credits pairing a vendor platform with its own fraud expertise for the results.