For decades, Financial Crime Compliance (FCC) has been burdened by manual processes, persistent alert backlogs, and inefficiencies that hinder risk detection. AI is now transforming this landscape by automating repetitive tasks, reducing the impact of false positives, and streamlining investigations. Proven AI agents are already deployed in leading FCC programs worldwide, cutting manual review efforts and freeing up millions of hours annually.
By shifting compliance teams away from low-value alert triage, AI enables investigators to focus on genuine risks while allowing financial institutions to optimize resources, eliminate backlogs, accelerate payments, improve onboarding, and drive business growth.
Recently, WorkFusion’s VP of Financial Crime Compliance David Caruso and Head of Product Marketing Kyle Hoback discussed how AI agents are helping to modernize FCC and drive operational efficiency to new levels.
AI Agents put many AI technologies to work
David pointed out that AI Agents combine many aspects of AI (ML, NLP, OCR, etc.) to purposefully augment FCC teams. That’s quite different than GenAI which focuses on generating content. Instead, AI Agents are designed to help people do their jobs with the content that already exists. So, the focus is placed on what is actually happening and where you can apply it in AML and FCC operations today. They are extremely pragmatic AI, reflecting what FCC teams need to build and operate programs, investigate cases, and file suspicious activity reports (SARs).
A prime example of this AI Agent pragmatism is seen in KYC (know your customer) operations. KYC work involves suspicious activity or sanctions detection, investigation, and reporting. You can categorize it into five major activities:
- Gathering information: customer records, third-party data like adverse media or background information on a company or business.
- Organizing information: Organized information is easier for ops team to work with
- Assessing the information: What do I have here? What are the relationships or their patterns of their activities?
- Reasoning: Attempting to decipher relationships and why they look
- Judgement / Decisioning: Deciding if something seems suspicious. Does it need to be referred for investigation? Can the matter be closed? Etc.
AI Agents help in all five areas, especially in the first three stages of gathering, organizing, and assessing information. To demonstrate, here is how the WorkFusion AI Agent named “Evan” for adverse media monitoring works:
AI Agent Evan operates by taking information from a standard adverse media screening tool and reviewing it to prioritize news articles and other information about your customers and potential customers. Evaluating relevance, demographic data, level of material significance, and other attributes, Evan is prebuilt with deep industry and prior FinCrime knowledge to fully investigate an entity (persons or organizations) with as little information as just a name. He highlights articles/information that indicate risk, passing those cases to a human analyst to review. Whenever Evan makes a decision, he provides detailed justification in written, auditable form.
The efficiency gains are massive
Consider that a typical adverse media search yields five results or more, each taking 10 to 20 minutes to read the article or other information and make sense of it. Evan automatically and rapidly reads them, identifies the people, parties, dates, amounts, and everything else that is relevant. On average, Evan optimizes a 20-minute human process down to 2 minutes, referring the riskiest cases to human counterparts via human-in-the-loop technology built into his standard functionality. In this way, FCC teams gain efficiency while maintaining ultimate control.
Transforming the transaction monitoring alert review process
David pointed out that the bulk of FCC and AML ‘people power’ goes into transaction monitoring (TM) which is often more complex than adverse media or customer due diligence. TM starts at simple transactions like structuring – wherein bad actors break down a large financial transaction into smaller, individual transactions to ‘fly under the scrutiny radar’ and avoid triggering AML alerts. TM then goes into high-risk transactions, such as those involving dozens of parties across multiple borders.
Whether a transaction is simple or complex, banks and other financial institutions look to WorkFusion AI Agents to solve the back half of the Detection>Alert Generation>Alert Review>Decisioning process. That’s because the front half of the process, Detection>Alert Generation is already addressed by many tools. But where the massive amount of resource spend occurs is on the back-end steps made up of labor-intensive Alert Review>Decisioning. That’s where all the people power has traditionally been dedicated, and that’s where banks and other FIs are looking to apply AI to make those people resources more efficient.
Their drive for FCC Ops efficiency is reflected in the new 2025 Dimensions survey results released by Celent. The survey, entitled Dimensions: IT Pressures & Priorities 2025, asked FCC leaders across FIs of all sizes, “What’s driving your technology spending and prioritization in 2025?”
The number one response, by far, was “AI for Efficiency.”
This targeted use of AI to gain efficiency aligns perfectly with the AI Agents by WorkFusion. Sure, they’re incredibly smart in their use of multiple forms of AI. But perhaps more importantly, they use that AI to automate for efficiency. With a WorkFusion AI Agent reviewing and making decisions around the overwhelming majority of alerts, banks and FIs can stop wasting resources and focus their people on resolving the most complex and high-risk situations, the ones that require nuanced human evaluation. As a result, Level1 analysts are no longer overwhelmed by massive volumes of false positive alerts and stay sharp to investigate the highly complex cases that matter the most for risk mitigation.
David summed up the value of WorkFusion AI Agents this way: “So, a more casual way to think about it is that every AML analyst or investigator now has a worker buddy, and it’s this sort of AI agent that’s doing a lot of these tasks for you.” It’s not that your screening system is producing fewer alerts than it did last month. It’s that the AI Agents consistently resolve the false positive alerts that used to drag down your operational efficiency.
Adverse Media Monitoring,AI,ai agents,AML,Compliance,FCC
