How to Address Present-day Sanctions Screening Pain Points with AI

Sanctions risk of financial institutions is evolving in line with the global social, economic and political changes. As seen in recent news, governments across the globe are increasingly relying on sanctions as an important measure for political foreign policy. They, as well as international organizations, implement preventive and corrective measures to prohibit illicit activity and control undesirable actions by certain high-risk countries, persons or groups.

For financial institutions, adhering to multiple sanction lists from various issuing countries and agencies is becoming a troublesome AML/CFT compliance task with process inefficiencies abound and risk heightened. What many of them lack are a sustainable screening framework and new-age screening tools. In this article, we would discuss the present-day challenges related to sanctions screening and a modern approach to address these challenges and build sustainable and scalable screening programs.

Why Sanctions Screening is Important for Financial Institutions

While all businesses in all sectors are mandated to comply with sanctions screening requirements, financial institutions, who work as channels of financial transactions, historically face increased scrutiny from regulators and enforcement actions have been more prominent on them. Therefore, financial institutions should have adequate controls in place to screen individuals and entities on a regular basis. They need to create a database of sanctioned individuals and entities and update them very frequently. In addition, they need to have tools in place to match their clients (both individuals and entities) with sanction lists, identify and stop unlawful activities, and report the same to relevant authorities. Failure in having adequate sanction controls and violation of sanctions would lead to enforcement actions including hefty fines. Penalties by the US Office of Foreign Assets Control (OFAC) reached a record US$1.3 billion in 2019.

What Makes Sanctions Screening Painful

The way how sanctions work is not uniform across the globe. It differs from country to country. There are sanction lists produced by countries as well as international bodies such as the United Nations and European Union. However, in a sanctions screening program, financial institutions need to compile information from various sanction lists and periodically update them. Further, they need to be watchful of the changes in sanctions programs to avoid risks. For example, OFAC updated 22 and 7 sanctions programs in 2020 and 2021, respectively, according to present official data.

The following are some of the key challenges of financial institutions with respect to sanctions screening.

• Multiple sanctioning bodies: Financial institutions may have to refer to lists produced by multiple sanctioning bodies depending on the territory of operation, currencies involved, the nature of business and international agreements.
• Daily updates: Financial institutions need to be watchful of any updates to their following watchlists on a daily basis. New entities are added to and removed from sanctions lists very frequently.
• Understanding sanctions: In line with global political and economic developments, the definition and scope of sanctions is broadening, and they are interpreted in different ways. Lack of clarity on sanctions is making it very difficult for financial institutions to effectively identify and manage risk. For example, customers who are not on a sanctions list but have some connection with a sanctioned individual or entity can also pose significant risk.
• Extended screening: At present, it’s not just customers that a financial institution should screen. They should have adequate controls in place to screen associates of clients, beneficial owners, and extended supply chains especially in geographies that have known links to sanctioned countries.

Technological Challenges in Sanctions Screening

Organizations are required to screen both their new and existing against multiple sanctions lists. Financial institutions either maintain in-house watchlists or subscribe to those provided by third parties. Subsequently, they check and match their customer and third-party databases in real-time or periodically with the help of certain tools for possible sanctions alerts. Possible matches are investigated and confirmed customers or third parties are blacklisted and reported. The objective of a sanctions screening program is not just detecting sanctioned customers and preventing them from doing transactions but it is also to avoid bad experiences to legitimate customers.

Recent changes in the sanctions space and the high volume of entries to be screened prompted financial institutions to move from rudimentary name matching models to rules-based screening tools. However, the volume of alerts generated for screening matches remained high with a false positive rate of more than 95%. These false positives are a drain on productivity as they take a lot of time and resources to remediate. This can lead to huge alert backlogs, high operational costs, poor customer experience and loss of business. With ineffective tools, there are also dangers of false negatives where designated entities slip through the compliance net, resulting in hefty fines.

The Way Machine Learning Augments Sanctions Screening Efficiency 

The primary reasons why existing screening tools remain inefficient and produce large false positives are:

• Inability to merge relevant data from multiple systems into a standardised structure
• limited consideration for secondary information such as date of birth, occupation, address and bank identification codes.
• Inadequate support for data in non-Latin characters
• Ineffective handling of name ordering, mis-spelling qualifiers, titles, prefix and suffix
• Lack of evidence-based alert review mechanism

In order to be effective, the technology used for sanctions screening should be easy to use and offer configurable risk-based settings, so that financial institutions can avoid over-screening and adjust screening criteria to match their risk appetite. By using machine learning, financial institutions will be capable of doing precision tuning their screening program to reflect the company’s risk exposure dealing with imprecise or inaccurate data to eliminate false positives.

As part of its award-winning Anti-Money Laundering Suite (AMLS), Tookitaki developed a Smart Screening solution leveraging advanced machine learning and Natural Language Processing (NLP) techniques. While addressing the above issues, the solution helps accurately score and distinguish a true match from a false match across names and transactions in real-time and in batch mode. In addition to screening against sanctions lists, the solution covers politically exposed persons (PEPs), adverse media and local/internal blacklist databases. The transaction screening feature triages and scores funds, goods or assets, between parties or accounts within a financial institution.

Tookitaki Smart Screening solution offers the following benefits to the customers:

1. More focus on alerts that matter

The solution offers a smart way to triage screening alerts by segregating them into three risk buckets – L1, L2 and L3 – where L3 is the highest-risk bucket. The highly accurate alert classification helps clients allocate time and experience judiciously and effectively address alert backlogs. Compliance analysts can focus on those high-risk cases (L3 and L2) that require more time to investigate and close. Meanwhile, they can close low-risk alerts (L1) with minimal investigation.

2. Better risk mitigation with reduced undetermined hits

Tookitaki solution uses NLP to process free texts and infers entity attributes like age, nationality, work-place title, alongside adverse media information, payment reference information or the stated purpose of the payment in a SWIFT message to derive vivid connection and accurately score all hits.

3. Superior screening accuracy with improved name matching

Tookitaki Smart Screening can handle typos, misspelling, nicknames, titles, prefix, suffix, qualifiers, concatenations, transliteration limitations and cultural differences for accurate hits detection.

4. Time/cost savings with faster implementation

Enabling faster go-live, the Screening solution comes with ‘out-of-box’ risk indicators across primary and secondary information of a customer for screening to accurately detect a true hit from several watchlist hits.

5. Low model maintenance costs

Too many lists with frequent updates have made screening more complex, prompting banks to introduce new rules and change thresholds. Tookitaki’s Smart Screening solution can self learn from incremental data and feedback to provide consistent performance over time.

6. Easy integration and flexible deployment

The solution has connectors to seamlessly ingest varied data points from multiple internal and external source systems and convert into a standardised format. Further, it provides API-based integration with primary screening systems, making the integration process easy, seamless and cost-effective. In addition, it offers on-premise and cloud deployment options.

7. Faster decisions with explainable outcomes

Tookitaki solution is equipped with an advanced investigation unit that provides thorough explanations for each alert and facilitates faster decision-making, reducing alerts backlog. Its actionable analytics dashboard for senior management helps monitor a bank’s sanctions risk across business segments, jurisdictions, etc. over a time period.

Recently, our AMLS solution went live within the premises of United Overseas Bank (UOB), one of the top 3 banks in Singapore, making us the first company in the APAC region to deploy a complete AI-powered AML solution in production concurrently to transaction monitoring and name/sanctions screening. By deploying AMLS, UOB could effectively create workflows for prioritizing alerts based on their risk levels to help the compliance team focus on those alerts that matter.

A complete revamp of existing sanctions compliance processes is imperative for financial institutions given that the international sanctions space is becoming more complex. It is time to embrace modern-era intelligent technology to enhance efficiency and effectiveness in AML compliance programs, establish next-gen financial crime surveillance and ensure robust risk management practices.

For more details into our Smart Screening solution, please contact us.