Practical AI in AML: measurable value across the lifecycle

Finextra will host an online webinar in association with NICE Actimize to examine where artificial intelligence delivers measurable value across the anti-money laundering lifecycle. The event will cover customer onboarding, know-your-customer checks, transaction monitoring, investigations, case management and suspicious activity report (SAR) preparation. A date for the webinar has been announced as to be confirmed.

The session will assess practical applications of machine learning, large language models and agentic AI, and how firms can move projects from pilots into operational AML programmes. Presenters will outline technical designs and operational changes needed to apply these technologies in live compliance environments.

Speakers will address specific stages of the AML process. They will discuss ways to speed onboarding, improve customer risk assessment for KYC, reduce false positives in transaction monitoring, automate parts of investigations and streamline SAR drafting. Use cases to be examined include adverse media research, screening optimisation, alert triage and case management automation.

The webinar will identify technical and organisational barriers that slow AI adoption. Presentations will cover methods for connecting risk signals across separate systems and workflows so compliance teams gain a single view of customer risk. Examples include combining transaction records, screening outputs and adverse media findings to prioritise alerts, and using natural language tools to summarise case files or draft narrative sections of SARs to speed investigator review.

Speakers will compare AI approaches. Machine learning models will be shown as the main tool for detecting anomalous transaction patterns. Large language models will be presented for open-source research, document review and report drafting. The panel will also examine agentic AI, defined as systems that execute multi-step tasks, and discuss its use in controlled automation of investigation workflows with human oversight.

Governance and regulatory compliance will be a central topic. Presentations will cover documentation of models and decision processes, requirements for explainability and transparency, and application of risk management frameworks. The webinar will describe testing protocols, ongoing performance monitoring and audit trails that teams can use to meet regulatory expectations.

The session will compare build, buy and hybrid approaches to acquiring AI capabilities. Discussion will cover the trade-offs between in-house development, vendor solutions and mixed models, and the procurement, integration and governance work each path requires.

Speakers will include examples where firms reported measurable outcomes such as improved detection rates, reduced analyst workload or shorter time-to-file for SARs. The programme will also outline practical steps for scaling AI efforts beyond pilots, including setting priorities, aligning projects with compliance objectives and integrating AI outputs into existing case management and reporting workflows.