Standardized Sanctions Screening Requires Standardized Sanctions Data

With the explosion in volume of sanctions against people, entities and nations issued by the United States and many other countries in recent months and years, AML Operations teams at banks of all sizes have recognized the need to scale their sanctions screening capabilities by standardizing the way they work and use data.

Standardizing operations around sanctions screening and AML seems fairly straightforward. You follow these seven steps (or a very similar set of steps) to achieve standardization:

1. Establish clear policies and procedures to determine how you will screen clients and transactions against sanctions lists, how you will review and escalate possible hits, and how you will treat true hits.
2. Standardize data in a consistent format so that you can easily screen it against sanctions lists
3. Develop a risk-based approach to reflect the reality that only some of your customers represent a higher risk level.
4. Perform ongoing monitoring in line with customers’ varying risk profiles.
5. Define an escalation process that specifies how you will act on possible sanctions hits, including investigation, internal reporting and notifying regulators and other authorities.
6. Document and record all sanctions screening activities to ensure smooth audits and to make the information easily accessible for any potential legal issues.
7. Regularly audit and review your operation to ensure compliance with current laws and regulations and ensure that your sanctions screening process is functioning as intended.

Yet, one of these steps has traditionally confounded banks’ AML Operations teams—standardizing data.

The challenge of standardizing data

Banks have long struggled with customer data that is incomplete, difficult to aggregate and hard to enrich with new and third-party sources of information. At the same time, AML Ops leaders know that standardizing their customer data can make automated screening processes highly accurate, efficient and easy to trust. But how do you actually ensure your data becomes standardized in a way that is both consistent and easy to screen against sanctions lists?

Enter IDP and AI

Intelligent document processing (IDP) and artificial intelligence (AI) are transforming data standardization at banks almost overnight. Leveraging IDP and AI in combination, your AML Ops team can dramatically improve the accuracy, speed, efficiency and trustworthiness of your customer data standardization process. With the resulting standardized data, your sanctions screening process will improve by leaps and bounds. “But again, how?” is the question AML Ops managers often ask, particularly for those who are new to AI.

Moving beyond the buzz of ChatGPT and other AI engines in today’s popular culture and financial news markets, AI can deliver very pragmatic results when used in a targeted, judicious manner as a part of IDP. Together, they automatically perform five vital tasks to standardize a bank’s data for sanctions screening. Here they are in the order that they occur:

Extraction of relevant data. IDP uses technologies such as Optical Character Recognition (OCR) and AI to extract relevant data from unstructured or semi-structured documents (e.g.., ACORD forms, tax forms, bills of lading, etc.). The extracted data can include customer names, addresses, identification numbers, transaction details and more. The NLP (natural language processing) subset of AI understands human language in context as well as nuanced language, enabling extraction of relevant data from text-heavy documents and even unstructured sources (e.g., emails, messages, etc.).

Data Validation. AI algorithms can be used to validate the extracted data. For example, they can confirm that an address is in a recognized format or that a series of numbers represents a valid identification number. They can also reconcile similar data across multiple sources and/or documents, such as a person’s name that appears in multiple documents/systems.

Data Standardization. Once the data is extracted and validated, IDP can transform it into a standardized format which is dictated by a sanctions screening team. For example, dates can be standardized to always appear as a 2-digit month, followed by a 2-digit day, followed by a 4-digit year. Similarly, addresses can be standardized as street number>street name>city>state>postal code.

Data Enrichment. AI can help to enrich the data by searching both internal databases and external data sources, such as commercial databases and the internet. For example, if a customer’s country of residence isn’t stated in their documents, but their address is, AI can use that address to determine the customer’s country of residence.

Data Classification. AI can classify the data into different categories, which can help with risk scoring. For instance, if a customer’s address is in a high-risk country or if the customer’s employment record shows them previously working at a sanctioned business, those can be flagged for further review.

The entire process from data extraction to data classification can be automated using IDP and AI. When this occurs, you can eliminate manual data entry, human errors and the expenses associated with having valuable people perform the process steps.

AI Digital Workers standardize your sanctions screening data—and more

Not only do AI Digital Workers leverage advanced IDP with AI to standardize any bank’s sanctions screening data, they also perform end-to-end sanctions screening roles. By incorporating ML (machine learning), they can search for more nuanced patterns of keywords, word omissions, combinations of names and context that may reveal previously-undiscovered sanctions risk and exposure. Over time, ML algorithms improve the accuracy of the data extraction and standardization process by continuously learning from past mistakes and successes.

WorkFusion’s AI Digital Workers are ready-to-hire automation solutions that have it all—AI/ML, IDP, NLP, OCR and automation. They are designed to work with documents, internal and external data sources, unstructured information, leading sanctions screening software and numerous systems. They make customer onboarding and continuous customer due diligence much easier in today’s growing sanctions landscape while dispositioning L1 alerts with greater efficiency and more comprehensive audit trails than teams of people.

As banks use AI Digital Workers to automatically standardize their sanctions screening data and sanctions screening processes, their highly trained people can be reallocated to more pressing compliance needs.