Hong Kong Records First Decline in Scam Cases as Banks Strengthen Fraud and Money Laundering Controls

Hong Kong recorded 43,212 scam cases in 2025, the first annual decline since 2019. Police attributed part of the improvement to closer cooperation with banks, expanded intelligence sharing and stronger frontline intervention.

Hong Kong reported its first annual decline in scam cases since 2019, offering an early indication that closer cooperation between law enforcement and the banking sector may be disrupting fraud and its associated money-laundering networks.

According to the Hong Kong Police Force, 43,212 scam cases were recorded in 2025, representing a 2.9% decrease from the previous year. Reported losses also fell from HK$9.2 billion in 2024 to HK$8.1 billion.

The figures were announced at the 2026 Outstanding Bank Staff Awards, where the Police recognised 16 employees and 19 banks for their contributions to fraud prevention and anti-money laundering.

Banks intercepted more than HK$2 billion

The Police’s Anti-Deception Coordination Centre and dedicated banking hotline intercepted 1,552 scam cases during the year. The value of intercepted funds reached HK$2.09 billion, an increase of more than 40% compared with 2024.

Frontline bank employees played a significant role. During 2025, 639 employees proactively assisted customers whom they suspected were being deceived. Their interventions prevented 595 scams from being completed.

Sixteen employees received individual awards for exceptional performance.

These cases highlight the continuing importance of human judgement. Automated monitoring systems can identify unusual transfers, but frontline employees may detect behavioural indicators that transaction rules cannot readily capture.

A customer who appears distressed, is receiving instructions by telephone or cannot explain the purpose of a transfer may require additional intervention even where the transaction is technically consistent with the customer’s account history.

Fraud prevention and AML are increasingly connected

Fraud and money laundering are frequently treated as separate compliance functions. In practice, they are closely connected.

Scammers rely on bank accounts to receive, divide and transfer criminal proceeds. These accounts may belong to complicit participants or to individuals recruited as money mules. Funds can then be moved rapidly through several institutions, converted into other assets or transferred overseas.

A successful intervention therefore does more than prevent an individual victim from losing money. It may also expose the accounts, devices, beneficiaries and transaction patterns supporting a wider money-laundering network.

For banks, this means that fraud intelligence should be incorporated into AML processes. Information obtained from scam reports and attempted transfers can support:

• Customer risk reassessments;
• Transaction-monitoring investigations;
• Identification of linked accounts;
• Suspicious transaction reporting;
• Account restrictions or enhanced monitoring; and
• Network analysis of potential money-mule activity.

Information sharing targets mule-account networks

Hong Kong has been expanding public-private and bank-to-bank information sharing to address the movement of criminal proceeds.

The Financial Intelligence Evaluation Sharing Tool, known as FINEST, enables participating banks to exchange information about customers, accounts and transactions that may be connected to financial crime. Its upgraded version is expected to include all 28 retail banks in Hong Kong.

A new legal framework also permits banks to share information where activity may indicate money laundering, terrorist financing or proliferation financing. This is important because individual institutions may see only one part of a criminal network.

One bank may identify the account that initially receives a victim’s payment, while another sees the funds being divided among secondary accounts. Without timely information sharing, neither institution may have sufficient evidence to understand the full pattern.

The Police reported that the number of identified mule accounts fell from 6,047 in the third quarter of 2025 to 5,781 in the fourth quarter, a decrease of 4.4%. Although one quarter does not establish a long-term trend, the decline provides a measurable indicator of potential disruption.

Scam alerts prevented further losses

Hong Kong’s Scam Alert programme uses analysis of suspected mule-account transactions to identify and contact potential victims before additional payments are made.

During 2025, the programme intervened in 4,060 cases and prevented losses exceeding HK$480 million. The number of interventions increased by 69.4% compared with 2024, while the amount protected rose by more than 150%.

This approach reflects a shift from investigating fraud after the money has disappeared to identifying victims and suspicious payment patterns in real time.

It also demonstrates how AML intelligence can support consumer protection. Transaction monitoring is not limited to detecting whether an account holder is laundering money. It can also identify when a customer may be sending money to an account already associated with suspicious activity.

Technology supports—but does not replace—human intervention

Hong Kong is also using technology to improve scam detection. The Scameter+ application introduced artificial-intelligence capabilities in October 2025 to identify fraudulent websites impersonating banks and investment platforms.

By the end of 2025, the application had been downloaded approximately 1.15 million times. Reported suspicious websites can be analysed, added to the relevant database and used to generate warnings for other users.

Technology can accelerate detection, but its effectiveness depends on current intelligence, appropriate escalation and human review. Banks should avoid treating automated alerts as a complete control. Employees must be trained to recognise the wider context of suspicious behaviour and know when to pause a transaction or escalate a case.

Cross-border cooperation remains essential

Many scams targeting Hong Kong residents involve criminal groups, payment channels or digital infrastructure located outside the jurisdiction.

Through the FRONTIER+ cross-border anti-fraud platform, participating authorities arrested more than 1,800 people and intercepted approximately US$20 million in suspected scam proceeds during joint operations in 2025.

Such cooperation is increasingly necessary because criminal funds can move across several institutions and jurisdictions within minutes. Rapid contact between banks, financial intelligence units and law-enforcement agencies can determine whether funds are successfully frozen or disappear through additional layers.

A model based on shared responsibility

Hong Kong’s 2025 figures do not mean that the scam threat has been resolved. More than 43,000 reported cases and HK$8.1 billion in losses remain substantial.

Nevertheless, the decline illustrates the potential impact of a coordinated response combining:

• Frontline employee intervention;
• Real-time transaction monitoring;
• Bank-to-bank intelligence sharing;
• Police and banking-sector cooperation;
• Proactive victim alerts;
• Technology-assisted website detection; and
• Cross-border enforcement.

For AML professionals, the central lesson is that fraud prevention and money-laundering detection should not operate in isolation. Banks that connect customer behaviour, transaction data, scam intelligence and information from other institutions are better positioned to protect victims and disrupt the financial infrastructure used by organised fraud networks.