Dive Brief:

  • HSBC, one of the largest banks worldwide, is partnering with a United Kingdom-based startup, Quantexa, to use artificial intelligence software to pinpoint money laundering activity in customer networks, according to a Quantexa announcement. The companies conducted a pilot last year and will broaden the software deployment in an effort to detect more illegal activity.
  • This is not the company’s first foray into AI applications to counter financial crimes. Last summer, HSBC partnered with another AI startup, Ayasdi, to automate anti-money laundering investigations, according to an Ayasdi announcement. The companies were able to reduce the number of investigations related to a finite number of cases by 20%.
  • HSBC joins other European banks using technology to fight financial crime, including the Royal Bank of Scotland, which is scanning small transactions for fraud, and Danske Bank, which is analyzing transactions in real time, reports Financial Times. Singapore’s OCBC also partnered with tech companies to improve detection and investigations of suspicious financial activity.

Dive Insight:

AI-enabled solutions have swept through the financial services market, turning tasks that once took multiple human workers hours and compressing them into seconds or minutes. As demand for technology solutions in the space rises, so does the money.

The global financial technologies market surpassed $959 million in 2016 and is expected to continue growing to $7.3 billion by 2022 — a compound annual growth rate of more than 40%, according to a MarketsandMarkets report.

Current investment is focused around data analytics, mobile, AI, cybersecurity and RPA, and emerging technologies drawing the most focus are AI, blockchain and biometrics and identity management, according to a PwC report. As more fintech companies put their innovation and solutions to use, it’s forcing financial institutions to work harder on updating legacy systems.

AI and machine learning hold applications beyond financial crimes. The tech can be used for data processing, regulatory compliance, credit assessment and trade optimization, to name a few.

One drawback, however, is a lack of transparency into how AI and ML models make decisions. Improvements to the tech improve transparency, but not understanding why an algorithm identifies something as a risk can prove risky for a company with compliance concerns.