Industry urges unified response to AI-enhanced fraud

Fraud experts say criminals are shifting from technical exploits to schemes that target human weaknesses, using AI tools and automation that make scams harder for consumers and institutions to detect. Interpol classifies financial fraud as one of the fastest-evolving transnational crimes and reports AI-assisted schemes are about 4.5 times more profitable than traditional methods.

Observers describe current scams as focused on social engineering and psychological tactics, often combined with deepfakes and automated messaging that imitate trusted sources. Industry participants say individual consumers and single firms can no longer carry the burden of prevention alone.

Responsibility for fraud prevention is spread across multiple regulators and oversight bodies, and no single authority currently has end-to-end oversight of cross-border incidents. At the same time, firms develop their own detection tools and reporting frameworks, which limits information sharing and slows industrywide response.

National privacy and data protection laws restrict what companies can share about customers and incidents. Regulators and jurisdictions set different reporting and monitoring requirements, which creates uncertainty about who should lead cross-industry initiatives. Smaller firms report resource constraints that make it harder for them to join complex information-sharing arrangements.

Proposed regulatory reforms include a common framework with minimum standards for incident reporting, classification and information exchange, paired with scaled or lighter compliance obligations for smaller firms. Another proposal calls for clear assignment of responsibilities among regulators so a single coordination mechanism can manage cross-border cases without imposing identical rules on all businesses.

Industry discussions include standardised processes such as a common taxonomy for fraud types, agreed formats for incident reports, secure data-sharing protocols or APIs, and shared threat feeds run by an independent body. Proponents say these measures could speed detection, reduce duplicated effort and produce more consistent metrics on fraud trends.

Limitations cited by experts include legal barriers to sharing customer data, antitrust risks, and the cost and technical burden for smaller firms in adopting common systems. Shared detection systems could amplify false positives across participants if thresholds are not carefully set. Questions remain about governance, funding, who controls access to shared data, and how to maintain impartiality.

A single, systemic non-profit platform would face cross-border legal differences and concerns about reputational or regulatory exposure from pooled data, as well as the need for sustainable funding and clear safeguards to protect competitive information.

A webinar hosted in association with Ecommpay will gather industry experts to discuss these regulatory and ecosystem changes. Panelists include Willem Wellinghoff, UK chair and chief compliance officer at Ecommpay, with Teresa Connors serving as moderator.