For the First Time, Nasdaq Is Using Artificial Intelligence to Surveil U.S. Stock Market
For the first time, Nasdaq is applying artificial intelligence on its U.S. stock market to detect irregular and potentially malicious trading activity. The newly launched initiative aims to strengthen and revolutionize market surveillance through machine learning and other artificial intelligence (AI) capabilities.
“The U.S. capital markets are the largest, most liquid financial ecosystem in the world and protecting our markets for retail and institutional investors alike is an important responsibility,” said Martina Rejsjo, Vice President and Head of Market Surveillance, North America Equities, Nasdaq. “This means constantly evolving how we adopt and leverage new technologies to better surveil trading activity. By incorporating AI into our monitoring systems, we are sharpening our detection capabilities and broadening our view of market activity to safeguard the integrity of our country’s markets.”
The new technology, which is currently pending patent with the U.S. Patent and Trademark Office, leverages specific AI capabilities for market surveillance, including deep learning, transfer learning and human-in-the-loop learning.
Deep learning allows computers to learn—with or without human oversight—from intricate patterns and hidden relationships in massive datasets, whereas transfer learning creates new models from old models. Meanwhile, human-in-the-loop learning lets analysts share their expertise with the machine efficiently, allowing them to concentrate their efforts on investigation and evidence curation. Presently, Nasdaq’s U.S. market surveillance team reviews more than 750,000 alerts annually, identifying unusual price movements, trading errors and potential manipulation.
“Artificial intelligence and machine learning have broad application abilities across our company – from providing a richer client experience, to predicting market trends or creating more sophisticated market surveillance capabilities,” said Michael O’Rourke, Senior Vice President, Head of Machine Intelligence, Nasdaq. “By implementing AI on our market surveillance system monitoring US equities, we are able to use unique approaches to learn patterns of market abuse, specifically, spoofing, and apply that knowledge to other marketplaces worldwide. AI will have crucial role in building the next generation of surveillance technology that focuses on adaptive detection models, which will empower surveillance teams with faster, smarter and more accurate monitoring capabilities to maintain market integrity.”
The project developed into a three-party collaboration between Nasdaq’s Market Technology business, its Machine Intelligence Lab and its U.S. market surveillance unit, together, creating machine learning technology for trade surveillance patterns. The initiative evolved into a year-long research and development pilot, designed to improve the company’s market surveillance functionality and detection of malicious activity with AI and transfer learning.
“By partnering with Nasdaq’s U.S. surveillance team, we are able to train our models based on the team’s experience in monitoring data directly from the trading engine of the world’s most dynamic market,” said Tony Sio, vice president and head of Marketplace Regulatory Technology at Nasdaq. “Through transfer learning, we have built a framework to provide those learnings to other marketplaces around the world. We believe this is a major step in the evolution of how we use artificial intelligence technology to maintain the integrity of the capital markets.”
The launch of this initiative also coincides with Nasdaq’s 2019 Global Compliance Survey, which found that firms are increasingly investing in artificial intelligence and machine learning capabilities to reduce false positives in trade surveillance. The survey data revealed that investments in this technology rose significantly, with 42% of respondents reporting that they have recently invested in it, and 65% planning to invest in it over the next 12 to 24 months.