From Chess Engines to Criminal Networks: AI’s Role in AML Today

For years, chess was considered the ultimate expression of human intelligence. It represented strategy, memory, pattern recognition, and intuition. These were all skills we saw as uniquely human. That belief was shaken in 1997 when IBM’s Deep Blue defeated world champion Garry Kasparov. It marked a pivotal shift in how we viewed the potential of technology. From that point on, artificial intelligence (AI) stopped being theoretical. It became real, measurable, and impossible to ignore.

Today, in 2025, the best chess player in the world is not a person. It is a machine.

Stockfish, the world’s leading chess engine, has an Elo rating of approximately 3640. For context, most casual players fall between 800 and 1300. A strong club player may rate between 1600 and 2000. Magnus Carlsen, widely regarded as the greatest human chess player in history, peaked at 2882. No human has ever surpassed that.

In chess, a 200-point Elo difference means the stronger player is expected to win 75 percent of the time. Stockfish has a margin of more than 800 points over Carlsen. The result is nearly always a loss for the human.

What we are seeing in chess is not isolated. Similar performance gaps are now appearing in industries that were once considered too complex or too human to automate. One of those industries is anti-money laundering.

AI Is No Longer Just a Tool. It Is the Benchmark

Artificial intelligence is redefining what expert performance looks like. In many sectors, AI is no longer just supporting professionals. It is setting the new standard.

In healthcare, AI models such as Google’s LYNA outperform radiologists in identifying certain types of cancer. These systems analyse millions of scans, detect minute patterns, and operate without fatigue.

In the legal industry, tools like Harvey AI review contracts, identify legal risks, and process complex case law in seconds. This allows legal teams to focus on judgment and strategy rather than review work.

In financial services, AI is used to screen customers, monitor transactions, and detect suspicious behaviour in real time.

AI is not replacing professionals. But in terms of speed, scale, and consistency, it is already ahead. And the gap is growing in areas where regulation and risk converge.

Criminals Are Using AI to Their Advantage

The same technology that strengthens compliance systems is also being used to defeat them.

AI’s darker capabilities are now being widely recognised across the compliance community. Generative AI is being exploited by criminals to create deepfake documents, forge identities, and automate complex laundering patterns. These tools can produce fake invoices, bank statements, and profiles that are nearly impossible to distinguish from legitimate ones. Combined with crypto assets and fast-moving virtual networks, this creates a highly scalable laundering ecosystem that is designed to outpace traditional controls (Learn more from FATFs Guidance on Virtual Assets and Virtual Asset Service Providers).

Other operations have gone even further, using AI to create fully developed fake businesses. These include product listings, customer reviews, and social media posts. Transactions appear legitimate, but the purpose is to clean illicit funds.

How AML Consultants Respond?

At Seamless AML, we help reporting entities stay ahead of these evolving risks. We are independent auditors and specialists in AML/CFT documentation and advisory.

Our services include:

• Independent AML audits that are structured, regulator-aligned, and defensible

• Customised AML/CFT risk assessments and compliance programmes based on your business model

• Practical documentation, including onboarding procedures, transaction monitoring guidelines, and escalation workflows

• Strategic advice on upgrading controls or integrating automation

• Ongoing support to keep your compliance approach aligned with new technologies and emerging risk typologies

  
  

Do we use AI in our own work? Ofcourse. We use it in a safe, controlled, and transparent way. It makes us more efficient and more accurate, which means we can spend less time on manual processing and more time on what matters. That also makes our services more cost-effective, and we pass those savings directly on to our clients.

The result? Smarter compliance support, delivered faster, without compromising on quality or security.

ChatGPT prompt: "create an image about chess, money laundering and AI"

Four Ways to Strengthen Your AML Programme Today

AI is raising the stakes, but it is also creating opportunities to improve. Here are four ways you can strengthen your AML response right now.

1. Refresh Your Risk Assessment

Outdated and generic risk assessments are a common weakness. Your assessment should be relevant, detailed, and reviewed regularly. It must consider current threats such as virtual assets, synthetic identities, and AI-generated fraud. New Zealand’s AML/CFT Risk Assessment Guideline makes it clear that your programme must be based on the risks you are likely to face.

2. Tighten Your Controls Around High-Risk Customers

Too often, enhanced due diligence is treated like a checkbox. But high-risk customers deserve more than just extra paperwork. Go beyond collecting documents. Ask better questions. Map out the customer’s activity before it happens. Understand their story, not just their structure.

Look at how their risk profile aligns with what you know about their sector, location, or payment behaviour. If it feels off, it probably is. Real risk-based compliance means slowing down when it matters, not when it is easy.

3. Make Your Compliance Programme Easier to Use

A good programme should be practical, not just compliant. If your team cannot understand or apply your policies, they will be ignored or applied inconsistently.

Strip out jargon. Simplify your procedures. Add examples that match the way your business actually operates. Build checklists and workflows that your staff want to use, not just have to.

This is what regulators mean when they talk about embedding compliance into your business. It should feel like part of your process, not a layer on top of it.

4. Use AI to Your Advantage

AI is no longer experimental. It is already being used in AML to monitor accounts, detect red flags, and streamline customer due diligence. If you are still relying on static rules or manual reviews, you are either missing patterns that matter or spending too much time on ones that do not.

Smarter tools can identify behavioural anomalies, pick up inconsistencies in customer information, and highlight risks that traditional systems might overlook. Some can even assess source of funds explanations or compare real-time activity against expected behaviour.

You do not need to automate everything. Start with one process. Use AI to make it faster, clearer, and more consistent. It gives your team more time to focus on what actually matters.

If you want to explore which tools are right for your business, get in touch with Seamless AML. We work with solutions that are practical, safe, and regulator-friendly.

AML, Chess, and Staying Three Moves Ahead

Whether you are a compliance officer, business owner, or reporting entity, the message is clear. Artificial intelligence is here. Criminals are using it. Regulators are responding. If your systems are standing still, you are already behind.

At Seamless AML, we believe compliance should be intelligent, efficient, and aligned with real-world risk. We use the best tools available, including AI, to deliver that. We are not afraid of change. We are built for it.

And if you ever want to talk about financial crime, or challenge me to a game of chess, I am always keen. Just do not ask me to play Stockfish.

Was this blog post written with a little help from AI? It sure was. Why? Because it’s fast, accurate, and never needs a coffee break. It helped shape the content, tighten the structure, and keep things engaging. How do we know it worked? You’re still reading. And if AI can do that, it’s worth paying attention to.

Let us outthink the threat together. If your risk assessment or compliance programme has not been reviewed in the past 12 months, now might be the time. Get in touch.