- AllianceBernstein has been building out a team focused on AI and data science since 2017.
- The team is led by Andrew Chin, AB’s head of investment solutions and data science.
- Here’s how AB uses AI to get an edge, save analysts hours of work, and improve risk management.
AllianceBernstein, a massive money manager, embarked on a seven-year mission to transform itself through AI and data science.
“At the time, I went to my CEO and said, ‘Look, we really need to rebuild our data science capabilities because our industry is changing and for us to be competitive in our industry,” Andrew Chin, who was serving as AB’s chief risk officer and head of quantitative research at the time, told Insider. He stated that the changes would be far-reaching, affecting everything from how AB collects data to how the organization is structured to how investment decisions are made.
Chin, now AB’s head of investment solutions and data science, would create an internal task force dedicated to machine learning and AI over the next several years. The move would save AB hundreds of thousands of dollars while also allowing its investors to gain a competitive advantage with the $704 billion assets they manage.
The mainstream release of ChatGPT and other forms of generative AI has reenergized Wall Street firms’ AI efforts this year. Other asset managers are making strides in AI and data science. BlackRock has also been experimenting with artificial intelligence to help employees do their jobs better and faster, with CEO Larry Fink predicting a 30% increase in productivity. Vanguard and Fidelity are both developing AI capabilities.
Chin revealed the asset manager’s AI strategy and use cases. Among the applications are using AI to detect differences in regulatory filings by portfolio companies, which can indicate poor performance, and using algorithms to detect changes in strategy shifts and manager terminations before news outlets.
Using AI signals to make investment decisions and time buys and sells
AllianceBernstein employs artificial intelligence to sift through more than 400 company reports and filings per day in order to identify potential risks in relevant companies, whether in AB’s clients’ portfolios or as a potential direct investment. Natural language processing, a type of AI that can process human language in a way that computers can understand, emphasizes this. The NLP is used to generate investment signals, which are alerts that can reveal poor performance or forecast positive movement.
One signal, for example, compares regulatory filings to look for differences in company strategy or management. Stocks underperform when there are a lot of differences because it indicates that the company is going through a lot of dramatic changes. “Usually, that means they’re dealing with difficulties,” Chin explained.
In one recent case, the NLP discovered significant differences in 10K filings of a retailer AB owned in one of its equity portfolios. The retailer had focused on increasing store count as a key component of its growth strategy, but AB’s AI tool discovered that this was no longer the case in the retailer’s 2023 10K. Chin’s team notified the retailer’s analyst. They determined that while the strategy shift would be viewed negatively by the markets in the short term, the fundamentals would remain strong in the long term.
Despite the fact that AB is a long-term investment firm, the asset manager maintained its position despite the stock price significantly underperforming. Chin noted that the stock recouped some of its losses about a month later.
AI is increasing productivity while improving risk management
AB uses natural language processing to monitor reports, news, and other written documents for language insights to measure potential company performance, in addition to comparing company filings. According to Chin, the level of complexity, or how difficult language is to understand, can indicate a company’s future performance. He went on to say that analysts and portfolio managers use these signals to generate ideas and time buys and sells.
AB also employs NLP to summarize new offering memorandums, which are legal documents distributed to potential investors in a private placement transaction. Because the documents are typically 300 pages long, it is difficult for analysts to evaluate these new opportunities on a daily basis. Analysts can quickly assess twice as many summaries for client portfolios using NLP, according to Chin.
AI benefits more than just front-office investors and analysts. AllianceBernstein is using AI and natural language processing in its back office, according to Chin.
In one instance, the firm’s operations analysts determine whether AB can invest in specific bonds or stocks for retirement accounts managed by the firm based on whether they comply with the Employee Retirement Income Security Act (ERISA).
The documents are about 150 pages long, and there can be up to 50 new securities issues every day, according to Chin. The firm employs natural language processing (NLP) to parse the documents, recommend whether AB should invest in the security, and provide a way to trace back to the supporting text. The tool makes a recommendation to the operations analyst, who makes the final decision. Because the decisions could be validated against the supporting text, the AI application allowed the teams to be 50% to 75% more efficient while improving risk management, according to Chin.