Experts Discussed AI Roles in Fraud Prevention

Panelists at a QUBE Events session advocated for combining artificial intelligence with human review to curb scams.

Updated on Sept. 23, 2026 in Financial Crime

Isometric editorial illustration featuring interlocking mechanical components and a translucent filter, representing AI-assisted financial fraud prevention systems.
Financial industry experts advocate for hybrid fraud detection models that combine AI analysis with human oversight to combat sophisticated scams. AI Illustration. Upload story photo >

Live Poll

Do you trust AI over human staff to accurately identify and prevent scams on your account?

Financial industry experts gathered to explore how AI can improve fraud detection systems. The discussion emphasized that human intervention remains essential for stopping emotionally driven financial scams.

Why it matters

Fraud prevention is shifting toward earlier intervention because traditional warnings often fail when customers are already emotionally committed to scammers. Breaking the momentum of fraudsters through unexpected delays is becoming a key defensive strategy.

A Bank of China (UK) customer lost £65,000 over six months in a romance scam that bypassed standard protections. Investigators currently use sequences of small test payments, such as £1 and £2 transfers, to identify potential money mules.

The players

Vivox AI

This organization specializes in artificial intelligence solutions for fraud detection and security.

Anne Markey

She is a financial industry expert who pioneered the use of machine learning for anomaly detection at UBS.

Bank of China (UK)

This is a major international financial institution providing banking services to customers in the United Kingdom.

UBS

This is a multinational investment bank and financial services company based in Switzerland.

The details

Fraud detection systems are increasingly analyzing behavioral shifts like unusual app activity and changes in voice cadence. While Confirmation of Payee systems can verify account names, they remain unable to confirm the authenticity of personal relationships, leaving gaps that require human review.

Timeline

  1. Anne Markey introduced machine learning for anomaly detection at UBS in 2012.

  2. A bank customer lost £65,000 over six months to a romance scam.

  3. Industry success metrics for fraud detection systems are projected for 2026.

Legal Context

The discussion on AI integration marks a significant evolution from the industry standard set by Confirmation of Payee systems. While those systems verify account ownership, the current shift toward behavioral AI seeks to address the vulnerabilities inherent in the existing legal and technical framework.

Increased use of behavioral AI may lead to more frequent security checks or temporary payment delays when apps detect unusual activity. These measures are designed to provide a necessary buffer against scammers by disrupting the momentum of unauthorized transactions.

The takeaway

Fraud detection is moving toward behavioral analysis to catch scams before they reach the point of no return. Customers should expect more friction in their digital banking experience as institutions move to proactively block potential mule activity.

Further reading

For more background on how institutions combat sophisticated schemes, visit the Financial Crime section.

Live Poll

Do you trust AI over human staff to accurately identify and prevent scams on your account?