Preventing scams can only take us so far. Following the money is the next step: Opinion

There is good news to celebrate. The Singapore Police Force (SPF) announced on Aug 26 that in the first half of 2026, the amount of scam losses fell 17.9 per cent to S$410.6 million compared with the same period in 2025.

This extends the positive trend from 2025, when losses from scams also fell 17.9 per cent from a record high of S$1.1 billion in 2024.

After years of grim, rising totals, the curve has finally reversed. Singapore’s anti-scam machinery – public education, the Scam Analytics and Tactical Intervention System, ScamShield, SPF’s Cyber Command and the enhanced security measures that financial institutions put in place – have proven to be effective.

The wins are real, but they are also front-loaded. Our defences have been concentrated at the door: spotting the scam, warning the victim and blocking the transfer before it happens.

Prevention is the right place to start, but it also has a natural ceiling. According to SPF, self-effected transfers – where victims move the money themselves – accounted for 80.8 per cent of reported scam cases in the first half of 2026.

When the act of sending money happens willingly and on the victim’s own device, detection and prevention can only reach so far.

Tracking the money’s trail

This is why the next fight is downstream: following the funds after they are sent and clawing them back. This is undoubtedly harder, but it is also where real leverage lies.

Following the funds means tracing it across the multitude of rails in the financial ecosystem at rapid speed.

The majority of scam proceeds still moves through fiat channels, but cryptocurrency matters out of proportion to its size, because it is where the money can quickly cross borders and disappear from domestic view.

A local dollar transfer is often the first leg, and from a mule account the funds are then swapped into crypto and routed out of the country.

Where does it go? Into an industrial scam economy across the region. The United Nations estimates that these scam operations may generate more than US$43.8 billion a year. They operate as an entire supply chain, and cryptocurrency is its settlement layer.

On-chain data shows that at the laundering layer, the money moves overwhelmingly in stablecoins – chosen for their low fees, deep liquidity and speed.

The proceeds then pass through purpose-built infrastructure. TRM Labs’ 2026 Crypto Crime Report found that Huione-linked platforms, the South-east Asian marketplaces that act as the region’s illicit clearing houses, have received more than US$96 billion in cryptocurrency since 2021.

But the same rails criminals exploit for speed are the investigator’s advantage. Many of these transactions happen on blockchains which are permanent, shared ledgers that can be traced, attributed and sometimes stopped.

Singapore’s own results point to the same.

The Anti-Scam Command recovered more than S$22.8 million in cryptocurrency in 2025, proving that when you follow the money on-chain, you can bring some of it home.

The role of private organisations

Public institutions and initiatives are already targeting the movement of money after they have left victims’ wallets, with some success.

What does this mean for compliance teams in financial institutions and other private organisations? For them, increasing how quickly and extensively they can track proceeds across borders and rails is of the essence.

Investing in tracing capability that does not stop at your entity’s own perimeter is necessary to effectively track the laundering activity downstream and is an advantage that firms should capitalise on.

Blockchain intelligence tools, for example, allow compliance teams to follow where the assets have moved across public blockchains, identify which other intermediaries or financial institutions might have been involved in the transactions, and work with them and the asset issuers to stop the proceeds from moving further.

Building predictive defensive artificial intelligence systems to better identify usable information is moving from a good-to-have to a must-have component of an effective compliance programme.

Criminal enterprises are already using AI to scale fraud, iterate laundering patterns and transfer funds faster than manual review can map.

Organisations must similarly leverage technological advances to help investigators and compliance teams outpace adaptive adversaries.

Foundation models can be built to take in the many individually weak, disparate signals of customers and their transactions to identify new patterns of illicit activity, and proactively flag such behaviour in data sets with billions of data points.

Most importantly, the collaboration between financial institutions, regulators, law enforcement and all other parties in the ecosystem such as cryptocurrency issuers needs to become as seamless as the illicit actors’ networks.

Real-time intelligence sharing and swift coordinated action across borders is the only way to tackle scam networks that move huge amounts in minutes.

International efforts are important

There has been encouraging movement in this direction. The Cyber Command runs a standing public-private partnership working from shared, near-real-time information and the Monetary Authority of Singapore’s Cosmic platform lets six major banks share intelligence on suspicious customers.

And even when the money leaves the country, international enforcement operations such as Operation First Light and the Frontier+ network, comprising 14 jurisdictions including Singapore, Malaysia, Indonesia and Thailand, have shown that cross-border cooperation can successfully bring it back.

Singapore has built one of the tightest front doors against scams, and it is paying off. The next challenge is downstream: following the money once it is sent, across borders and across rails, and taking it back. This fight can be won, but only if we all work together to move as fast as the money does.

The writer is head of compliance advisory, Asia-Pacific, at TRM Labs

The commentary is based on the writer’s own experiences, observations and argument. AI tools were used for research and drafting. The writer remains fully accountable for the commentary’s accuracy, originality and final form. 

Source: The Business Times © SPH Media Limited. Permission required for reproduction.

ADV: Out now: Information and Communications Technology Law (2nd Edition) Committee formed to strengthen safeguards for children in religious settings, report due end-2026