Use GenAI to fight financial crime with confidence

Enhance the efficiency and effectiveness of your AML and sanctions compliance

Anti-money laundering (AML) and sanctions programs are growing more complex. Regulatory expectations are evolving, enforcement actions are becoming more severe and cryptocurrencies are creating new avenues for financial crime.

New technologies, including generative artificial intelligence (GenAI), machine learning and network analysis tools, are disrupting traditional ways of detecting and preventing money laundering, terrorist financing and sanctions violations. We’re paying close attention to GenAI-powered tools that can identify transaction monitoring capabilities and create new rules based on even more meaningful scenarios—helping financial institutions better pinpoint incidents of suspected money laundering. Additionally, smarter string-matching processes can reduce false positives and produce more productive sanctions, politically exposed persons and adverse media flags.

These powerful technologies can help financial institutions detect suspicious transactions more effectively. But without the right implementation plan, organizations can quickly find themselves confronting a mounting backlog of alerts as well as tough questions from both regulators and internal risk management teams responsible for model validation.

Conversely, we’re seeing financial institutions successfully use these emerging technologies to meet new regulatory requirements and combat financial crime with greater efficiency and effectiveness, enhancing their customers’ experiences. By pursuing innovation and growth, they’re freeing up staff for more value-added work—increasing employee engagement and providing more depth to their investigative processes.

Getting there requires comprehensive AML and sanctions programs that incorporate several key considerations.

Prepare to use GenAI for AML and sanctions compliance

Coordinate your first- and second-line data requirements

The ability to generate scale and value from technologies such as GenAI depends on how well financial institutions make use of their data. Gaps in know-your-customer (KYC) records—such as blank client birth dates or occupations recorded in free-form fields—can make working with GenAI and machine learning models more challenging.

Overcoming these hurdles starts with better communication between the first and second lines of defence. This helps these teams agree on what data your organization’s AML and sanctions programs need to function and the format it must follow. It can also build trust with regulators, who want to see the first and second lines of defence working in conjunction.

Build new models with model and data governance frameworks in mind

Some financial institutions are hesitant to adopt advanced AML and sanctions solutions because of the heightened governance requirements on their evolving underlying models. In other cases, business leaders admit their excitement is getting ahead of proper safeguards: 73% of Canadian respondents to our Global Digital Trust Insights survey said they’d personally feel comfortable launching GenAI tools in the workplace without any internal policies for data quality and governance.

Gaining an outside perspective can help you adopt GenAI and machine learning governance frameworks with confidence. This lets you explain to regulators how you're using and monitoring the models that underpin these technologies while continuing to innovate and evolve your AML and sanctions programs.

For example, we recently used an unsupervised machine learning clustering technique to help a financial services organization better spot low- and medium-risk customers engaging in cash-intensive transactions or complex relationships indicative of money laundering. We built a model that parsed out customers who conducted transactions in similar ways and identified the outliers within these groups, as opposed to finding outliers within entire segments. We then layered other risk-based factors into the model to recognize previously uncovered abnormal behaviour—increasing the organization’s ability to detect money laundering through more targeted monitoring.

Help employees navigate change by making upskilling part of your business readiness plans

Large numbers of employees require training before new systems used for transaction monitoring, customer screening, watchlist management and other AML and sanctions programs go live.

Thinking beyond investigators and other primary users by considering more peripheral users, such as employees who fine-tune the system rules, mitigates the risk of implementing new technologies. It also helps your employees analyze the information these tools generate, resulting in more effective risk ratings and transaction monitoring.

For example, our financial crimes team recently led the initial technology design and implementation of a new system based on technology from one of our key Alliance relationships for a large financial institution. In the process, we spotted an opportunity to upskill the organization’s employees to give them a stronger understanding of the new system, the results it produces and how it provides them with more time to conduct their investigative process so they can better spot potential threats.

This upskilling imperative is likely to grow. Our 2023 Global Hopes and Fears survey found many Canadian employees are keen to use GenAI to improve their productivity at work and learn new skills. Employee upskilling augments your wider business readiness plans, which also includes having the right people, capabilities and procedures in place.

Consider, for instance, how detecting suspicious transactions with greater effectiveness can lead to more alerts requiring prompt adjudication. Accurately estimating this impact in advance lets you line up the right internal staff and external support, including managed services providers that can quickly scale resourcing up or down, to manage your alerts.

A more effective and efficient way to fight financial crime

Financial institutions that successfully use GenAI, machine learning and other emerging technologies to fight financial crime can run their AML and sanctions programs more efficiently. They free employees to investigate previously overlooked risks, helping their organization spot suspicious activity with greater effectiveness.

This can prevent money laundering and terrorist financing from happening in the first place—helping financial institutions build trust, enhance their customers’ experiences and create a more safe and secure society.