Stay up to date with free updates
Just sign up Fund Management myFT Digest – delivered straight to your inbox.
Asset managers are increasingly using artificial intelligence to guide investment decisions, track the habits of portfolio managers and identify money-making opportunities.
JPMorgan plans to expand its use of generative AI tools later this year to flag questionable decisions by portfolio managers, such as potentially premature sales of high-performing stocks, executives at the firm told the Financial Times.
The tool, called “Moneyball,” shows portfolio managers “how they and the market have performed in similar situations, helping them correct biases and improve their processes,” said Christian West, head of investment platforms at JPMorgan Asset Management.
Other fund managers are using AI to complement human analysts, identify litigation funding targets and explain allocation decisions to investors.
These various initiatives signal that the AI wars in asset management are shifting away from paper-heavy compliance and marketing tasks toward helping investment professionals make smarter decisions.
JPMorgan’s tool is part of the $3.2 trillion asset manager’s Spectrum portfolio management platform, which is being developed using about 40 years of data, and is being developed as a pilot program that will be rolled out to a wider range of portfolio managers later this year.
Boya Investment Management has already deployed virtual analysts to monitor potential risks in stocks, complementing the $331 billion firm’s human research staff. Portfolio managers have access to a dashboard that displays security reviews by human analysts side-by-side with AI feedback, such as red flags for stocks.
So far, Voya’s AI analysts have made a high ratio of right and wrong decisions, making their warnings “high-value signals,” said Gareth Shepherd, co-head of machine intelligence at Voya, who likened the process to a pilot and co-pilot reading signals from an airplane’s flight-management system.
“The flight management system assists the pilot in making decisions, but the final decision rests with the pilot,” Sheppard said.
Legalist, a $1 billion hedge fund focused on litigation finance, is using a proprietary AI search tool called “Truffle Sniffer” to sift through the vast amount of civil litigation to find attractive investment opportunities.
This “sniffing device” scans court records for signs of favorable outcomes, such as friendly judges, case classes and pretrial rulings that could indicate a particularly favorable case.
“We’re looking for cases where there are clear signs of success, but where no money has yet been collected,” Legalist co-founder Eva Shan told the Financial Times.
In some cases, AI has a significant say, such as with Craft Technologies, an AI-powered exchange-traded fund backed by South Korean conglomerate LG and SoftBank.
The firm’s LQAI ETF, which launched in November and utilizes a proprietary AI stock-picking tool, has evolved to incorporate monthly AI-generated holdings reports. Recent AI-generated reports explain the firm’s decisions to favor certain companies and sectors and sell others.
“As a portfolio manager at LQAI, I have increased my exposure to resilient, technologically advanced companies (such as Google parent Alphabet) while slightly reducing my exposure to struggling traditional media companies. This reflects a cautious yet optimistic approach to capitalize on growth opportunities amid financial volatility,” the AI-generated holdings report said.
Despite these advances, some remain skeptical about AI’s ability to boost long-term profits for asset managers.
Veteran portfolio manager David Giroux, who runs the $59 billion T. Rowe Price Capital Appreciation Fund, argued that most of the AI-focused intellectual capital in asset management is directed toward finding short-term advantages, rather than the harder task of forecasting potential returns years into the future.
“I think it’s highly unlikely that AI will be able to eliminate those inefficiencies,” Giroux said.
