AI has changed fraud. Now we need to change how we fight it.

Shufti August 24, 2026 6 minute read

Back

The shape of fraud has changed

Most of your defense systems were built to defend against attacks coming at human pace, which made it easier to track down and take care of.

AI has changed the way we see fraud now. AI has made it easier for continuous, machine-generated frauds arriving at every point in your customer journey.

We need to change the way we see and fight fraud

The shape of fraud has changed

The core issue isn't a surge in the overall volume of fraud.

While topline financial losses appear relatively unchanged, the nature of the threat itself is rapidly transforming underneath which is a shift that can easily go unnoticed.

Why most defenses are still solving last year's problem

Legacy defenses are now attempting to address an entirely shifted paradigm rather than failing at their original purpose. The nature of the threat has evolved beyond what these systems were designed to handle.

From verification to intelligence

A subtle transformation is taking place within fraud departments, mirroring the evolution that credit risk underwent fifteen years ago. In this new paradigm, verification serves merely as a single component of the data stream rather than the final conclusion.

Using AI inside a regulated business

AI acts simultaneously as the attack surface and the defense framework. While attackers leverage it for synthetic identity generation or deploying novel threat vectors, defenders utilize it for real-time anomaly detection and adaptive responses to emerging threats.

Although both sides gained access to the technology at nearly the same time, only the defending side operates with a change advisory board.

THREE SHIFTS OVER THE NEXT 24 MONTHS WILL DEMAND

Three shifts over the next 24 months will demand

This is a roadmap, not a forecast

Continuous Auth

Verification is the start, not the end

Every login, transaction and high-risk action gets a fresh risk read

Signal Fusion

The orchestration layer is the product

Identity, behavioural, device and contextual signals in one decision

Adaptive AI

AI is the operational layer, not a feature

Built for the regulator and the attacker, simultaneously

Meet the Panel

Richard Bromley

Director Head of Fraud & Disputes Risk, Monzo Bank

Richard Bromley is Director and Head of Fraud & Disputes Risk at Monzo, with previous fraud leadership roles at Revolut, Aldermore, and Worldpay. Specialising in fraud and anti-bribery and corruption across banking and payments, he holds an ICA Diploma in Financial Crime Prevention and is a frequent keynote speaker as well.

Frayyam Asif

Chief Technology Officer, Shufti

Frayyam Asif is the Chief Technology Officer at Shufti, where he leads the engineering and IT teams building the company's AI-powered identity verification and compliance platform. He brings extensive experience in full-stack development, DevOps, and software quality, and previously served as Chief Development Officer at Programmers Force. Faryam leads a team of 29+ engineers delivering Shufti's core technology infrastructure.

Oonagh van den Berg

Chief Evangelist, Tangos AI

Oonagh van den Berg is the Chief Evangelist at Tangos AI and the founder of Virtual Risk Solutions and RAW Compliance. A financial crime specialist with over 20 years' experience, she has held senior AML and compliance roles at Citi and HSBC and now advises banks, fintechs and regulators on financial crime risk, governance and the practical use of AI in compliance. She is a regular speaker and trainer, known for making complex regulation usable for the teams who have to apply it.