Financial services providers that slack on regulatory compliance and fail to safeguard their operations against money laundering, terrorist financing and other criminal activities may face damaged reputations and significant fines. Compliance failures are prevalent worldwide: Approximately $26 billion worth of fines were levied against banks for AML, KYC and sanctions noncompliance between 2008 and 2018. A report found that the U.S. imposed a full $23.52 billion — 91 percent — of those penalties, while European regulators demanded $1.7 billion and the Middle East levied $9.5 million.
FinTechs could face these same financial pains as regulators increasingly demand that they follow the compliance rules to which FIs must adhere. The People’s Bank of China announced in March that it plans to create rules for regulating and securing the FinTech sector, for example.
FIs and FinTechs increasingly encounter new forms of fraud as they expand their digital operations, making it all the more important that they have strong risk assessment and compliance systems in place. This month’s Deep Dive examines the struggles and strategies involved in securing the FinTech and digital banking space and how AI may be able to help.
Financial Companies’ Security And Regulatory Obligations
Financial sector players must guard against all forms of money laundering and other criminal activities. One of these is “smurfing,” a technique that allows fraudsters to get around banks’ protections and transfer large quantities of illicit funds by dividing the money into smaller deposits placed into many different customer accounts. Banks are not required to report the deposits to the Internal Revenue Service (IRS) if each is below $10,000, making this kind of attack hard to detect.
The growing prevalence of cryptocurrencies is also complicating the finance sectors’ security efforts. These currencies typically provide anonymity and quick international transactions, features that can facilitate money laundering and terrorist financing. Regulators are increasingly taking note of such problems, with some seeking to improve AML and anti-tax evasion efforts by prohibiting anonymous crypto transactions. Australia began regulating digital currency exchanges and requiring compliance with AML and counterterrorism financing (CTF) rules in April 2018, and France’s Finance Committee raised a call to ban the trade or distribution of digital assets that enable anonymous dealings in March 2019.
FIs and FinTechs alike must stay vigilant and keep up with new risks and regulations as the threat landscape shifts. They must also ensure that they do not sweep up legitimate transactions in their fraud fighting efforts, as this may introduce customer frictions, and the resultant investigating and filing of suspicious activity reports (SARs) on false positives might drain
resources. A team of analysts can only handle so many potential fraud cases at a time, after all.
Can AI Support Digital Banking’s AML Efforts?
Some solution providers and observers in the space argue that many FI and FinTech AML and KYC woes stem from the struggle to acquire, sort through, fully comprehend and identify patterns in high volumes of data. Companies can experience KYC blind spots when onboarding new customers if they do not draw on the correct informational databases, many of which help them recognize important links the applicants may have to criminals, politicians, public figures or terrorist organizations. Those details need to be effectively analyzed, and FIs and FinTechs must continue to monitor transactional information to ensure no issues emerge. Solution providers claim artificial intelligence- (AI) and machine learning- (ML) based systems could aid in this by processing greater amounts of data at faster rates.
Automation tools are expected to take on the more repetitive tasks involved in AML and KYC compliance, as well as those that entail high levels of data-crunching. Leveraging automation will likely produce fewer false positives and reduce the time it takes FIs and FinTechs to investigate red flags. ML is also capable of keeping defenses up to date by adapting to emerging forms of fraud.
Effectively using AI still requires tackling some pain points, however. Some would-be adopters may question if the technology is advanced enough to make accurate, consistent assessments. Others might find it burdensome to invest in new software and systems and change existing processes. Solution providers can alleviate such concerns by avoiding bias in AI training, establishing performance metrics for the software and assisting with gradual solution implementation — all of which helps avoid the disruption caused by quickly and significantly overturning existing business models. Providers should also ensure that clients understand the AI software’s abilities and limitations. This allows the latter to make well-informed judgment calls based on the tools’ assistance.
Money laundering, tax evasion and other types of financial fraud pose significant and evolving threats that active participants in the financial industry must contest. FinTechs looking to expand their services and customer bases, and FIs seeking to remain relevant, must find cost-and time-effective ways to identify and neutralize misbehavior or risk customer defection and regulator fines. Leveraging the right technology can help these companies better support their fraud analysis teams and make the most out of their resources.