Artificial intelligence (AI) has industrialised fraud, and financial institutions can no longer rely on isolated detection to stop it. Criminals in the UK stole £1.28 billion through payment fraud in 2025, with Authorised Push Payment (APP) scams up nineteen per cent. The only proportionate response is collective intelligence: consortium data and analytics and real-time information sharing that provide the broader context needed to identify and disrupt criminal activity at scale.
The numbers in UK Finance’s Annual Fraud Report 2026 tell a story that impacts every financial institution in the country. Payment fraud increased four per cent in 2025, the second consecutive year of growth. Behind that headline sits a more revealing shift: unauthorised fraud losses fell five per cent to £703.4 million, while APP scams climbed nineteen per cent to £576.4 million.
Fraud is migrating from the unauthorised side of the ledger, where a bank can still intervene before money moves, to the authorised side, where the customer has been manipulated into sending the payment themselves.
How has AI changed the fraud threat?
AI has industrialised social engineering, equipping criminals to clone voices, fabricate identities, and generate convincing communications at a scale and speed that traditional controls were never built to absorb. Nasdaq Verafin’s 2026 Global Financial Crime Report found that ninety per cent of financial crime professionals surveyed reported an increase in AI-driven attacks over the past two years. The speed at which AI-enabled fraud is reshaping the threat landscape is unprecedented.
As Nasdaq Verafin noted in its analysis of the UK Finance Annual Fraud Report 2026, criminals are increasingly using AI-powered tools to gather information, create convincing communications and scale their operations more efficiently. The result is fraud that is becoming faster, more personalised and more difficult to detect. Key shifts include:
- Scams have evolved from easily identifiable phishing attempts into highly personalised attacks that can unfold over weeks or even months
- Fraudsters increasingly leverage deepfakes, voice cloning, synthetic identities and hyper-personalised phishing
- Fraud is becoming more scalable, more persuasive and more operationally demanding
Notably, some legacy attack types like remote access fraud and malware actually declined. Those declines indicate a displacement as criminals migrate toward lower-friction tactics where manipulation is harder to detect.
Why can’t institutions fight fraud alone?
No single institution can see enough of the wider pattern to act in time. If the threat is AI-powered and coordinated across institutions, then a defence built on isolated transaction scoring cannot keep pace.
Consider the APP scam in particular. Conventional payment fraud controls profile only the customer within a single institution; they look at the sender. But as risk migrates to the receiving side of a transaction, that inward-looking view is increasingly blind to the real danger: where the money is going, and who controls the account it is going to. A receiving account that shows the trademarks of money mule activity is invisible to a bank that can only see its own side of the transaction.
This is the core problem as criminals increasingly operate on a network scale. They share playbooks, infrastructure, and intelligence across institutions and borders. A fraudster who has targeted one bank today will target another tomorrow, often using the same mule accounts, the same device fingerprints, the same stolen identities. Each institution sees only its own fragment of that pattern, and by the time the fragment becomes a confirmed loss, the criminal has already moved on.
What is consortium intelligence in fraud prevention?
Consortium intelligence is built on anonymised, cross-institutional data that enables financial institutions to detect fraudulent activity no single organisation can see alone. This networked data and intelligence fuels consortium analytics and real-time information sharing, providing institutions with key insights.
This is the principle behind Nasdaq Verafin’s consortium approach. A network of more than 2,800 financial institutions, including global tier 1s, contributes anonymised, cross-institutional intelligence spanning more than 850 million counterparties. That consortium intelligence allows financial institutions to uncover fraudulent activity quickly and effectively, across the entirety of a transaction, in ways no single institution can see alone. Critically, it does so without sharing personally identifiable information (PII): institutions contribute to and benefit from a shared view of risk while protecting customer privacy.
How does consortium analytics stop APP scams?
Consortium analytics can stop APP scams by profiling the receiving account as well as the sender, helping institutions distinguish a low-risk account from a potential money mule before the payment is released. The result is more fraud caught, fewer legitimate transactions blocked, lower operational costs, and less loss to both the institution and its customers.
This kind of intelligence also extends beyond fraud. The same consortium data helps institutions surface high-risk entities both inside their own institution and beyond it, including shell companies, mule networks, and counterparties previously flagged as high risk by another institution, reducing false positives, accelerating investigations, and improving the decisions that matter most.
What should financial institutions do now?
AI is accelerating the scale and sophistication of fraud, but it is also revealing an important truth: no single institution has a complete view of the threat landscape.
As criminal networks become more interconnected, financial institutions will need to rely on the same principle that makes those networks effective: shared intelligence. At Nasdaq Verafin, we have built our financial crime management platform to deliver consortium insights and reveal a more accurate picture of risk. Connected criminal networks demand a connected defence.
In that environment, consortium intelligence is the solution that can allow financial institutions to outpace criminal networks.
FAQs
What is Authorised Push Payment (APP) fraud?
APP fraud occurs when a customer is manipulated into authorising a payment to a criminal. Unlike unauthorised fraud, where a transaction is made without the customer’s consent, the customer themselves initiates the payment, making it far harder to detect and stop.
How is AI used in fraud?
Criminals use AI to clone voices, fabricate identities, and generate convincing phishing communications at scale. According to Nasdaq Verafin’s 2026 Global Financial Crime Report, ninety per cent of financial crime professionals reported an increase in AI-driven attacks over the past two years.
What is consortium analytics in fraud prevention?
Consortium analytics is the use of shared, anonymised data across multiple financial institutions to detect fraud patterns that no single institution can see alone. Nasdaq Verafin’s consortium spans more than 2,800 institutions and 850 million counterparties, without sharing PII.
How does consortium analytics help with APP scams?
By profiling the receiving account as well as the sender, consortium analytics helps institutions identify potential money mule accounts before a payment is released, catching more fraud while blocking fewer legitimate transactions.
About the Author
Kamlesh Harry
Principal Strategic Advisor of Fraud Solutions
Kamlesh Harry currently serves as Principal Strategic Advisor of Fraud Solutions at Nasdaq Verafin, leading the global financial crime prevention strategy. During his career built at global tier 1 banks across South Africa, the U.S. and the U.K., he has helped financial institutions identify and mitigate financial crime risks while advancing payment fraud risk management, cyber security and business operations. A passionate advocate for raising awareness and education amongst students and vulnerable communities, Kamlesh invests personal time volunteering with the local law enforcement in the fight against financial crime. He also actively promotes fraud, financial crime and cyber risk awareness across markets, fostering skills to build a safer digital ecosystem. He is a Certified Information Security Manager (CISM) as accredited by ISACA.
