
Artificial intelligence systems for trading, factor identification and portfolio construction have not been able, to date, to communicate with each other effectively. But the technology is constantly advancing, and it seems only a matter of time before this AI synergy is reached.
When this happens, will AI replace human asset managers and trigger an overhaul of financial risk?
Aaron Brown
Hedge fund executive Cliff Asness says AI is becoming “annoyingly good” at doing parts of his job. AI deployed by his firm AQR Capital Management is combining investment factors to build market-beating portfolios, something that used to be Asness’ specialty. “AI’s coming for me now,” he told Bloomberg Television in a recent interview.
I worked for Cliff for 10 years, and we’re friends, so I care whether or not AI comes for him. But everyone should care because AI is not just coming for Cliff, but for the entire investment management business. That represents trillions of dollars of market cap that could evaporate, potentially taking thousands of buy-side risk management jobs with them.
Replacing human asset managers with computers will reshape financial risk in fundamental ways. But it will also open massive new opportunities for both individual risk managers and businesses.
Transformative AI: The Dream of Self-Driving Portfolios
Almost exactly seven years ago, Asness had expressed skepticism to the Financial Times, saying that big data and machine learning were dangerous because they found too many spurious patterns, and even genuine patterns were quickly competed away in the markets. However, like a good portfolio manager, he hedged his bets, saying, “We’re feeling our way. If our first few experiments bear fruit, we’ll do more of them. If we find out we’re good at this, it will become a bigger part of AQR.”
Going back a further seven years to 2010, I recall the early enthusiasm for AI in quantitative investing. Breakthroughs in AI algorithms and improvements in computer processing caused a gold rush mentality among many investment managers, and quants seemed well positioned to be the first to the motherlode, as they had the training and skills to understand and apply AI. Initial results, though, were disappointing — not terrible, just not the kind of improvements that technophiles and science-fiction fans had hoped for.
But the last seven years — conventionally dated to the 2017 publication of “Attention Is All You Need” by Alphabet Inc. researchers — have changed the picture dramatically, and the dream of full self-driving portfolios seems within reach.
There are three main steps in quant investing: identifying factors such as value and momentum that predict future returns; combining signals from those factors into optimal portfolios;’ and executing trades to keep the actual portfolio optimally close to the optimal one. From 2010 to 2017, AI proved unsatisfactory at identifying factors or combining signals. It was helpful at trading, providing the sort of improvements we got from standard methods – but not a quantum leap.
In the last seven years, from 2017 to 2024, as Asness said, AI has finally begun to pay off in factor identification and portfolio construction. But that’s not the big story. The most exciting advance on the horizon, the one that might finally realize the science-fiction dream, is AI researchers have figured out how different types of AI systems can communicate. The synergy from combining three different AI systems should be much greater than the individual advantages in each step.
For example, the trading AI should be sending news about soft prices or erratic volumes in securities back to the factor-identification and portfolio-construction AIs, which in turn should be sending insights to each other and the trading AI. This was one of the key advantages of old-style hedge fund managers who did their own research, portfolio construction and trading. But in large quant hedge funds today, those tasks are done by different specialists.
We’re not quite at the point of breaking the curse of the Tower of Babel. Although different types of AI systems can communicate, it’s still more of a pidgin sign-language than a fluent and precise exchange of information.
Moreover, even the best AI systems cannot rid themselves of occasional gross errors — the kind no human would make. And even one of these can destroy the effectiveness of the entire process.
That’s why we still have humans examining the output of each AI system before feeding it as input to the downstream system (from, say, the factor-identification system feeding signals to the portfolio-management system) and why upstream communication (from, for example, the trading system to the factor-identification system) is restricted or forbidden.
But these problems now seem solvable with some algorithm tuning and more computer power, issues for workaday quants rather than anything requiring breakthroughs by geniuses.
In addition to the three steps above for the core investment process, in 2010 we hoped for three indirect aids: AI coding, monitoring and explaining. Coding was the first home-run success of AI in quant finance: no one needs to code anymore, as AI does it much better, faster and cheaper. But we’ve had no success at monitoring tasks – such as having the AI issue alerts that some data looked fishy or that some factor was not behaving as expected – and none at explaining events – like why, for example, some price relation in the market had broken down. I’m seeing progress in these areas as well, although slower than in core quant investing.
Rethinking Investment Management
If all this works, we may see the end of the investment management business as we know it. Instead, each person could have a personal AI that never sleeps or has its attention wander. These AIs could trade with each other with no need for human intervention, combining personal information — tax situation, financial goals, income prospects — with superior market knowledge.
Imagine checking your phone in the morning — or perhaps putting on your smart glasses or listening to a verbal summary — to check your AI financial assistant. It will have been running constantly, tracking any expenditures (like subscriptions) and amortizing expenses like taxes and revenue. What’s more, it will move your assets around to maximize interest and gains while maintaining precisely calibrated liquidity and risk, finding opportunities that might interest you, and updating your long-term plan.
There would be no more tax filings, or searching for mislaid financial information, or worrying about forgotten funds or periodic charges. There’d also be no need to read financial news, or to wait on hold on the telephone, or to negotiate confusing websites – all while making errors and missing opportunities.
Undoubtedly, existing investment managers will try to charge fees for proprietary AI financial planning and investment bots, and fintech companies will offer their own versions, claiming superior technology rather than specialized financial knowledge. I suspect most people will let big companies like Alphabet Inc. or Amazon give them a high-quality free version, in return for the massively valuable information these companies will gain from aggregating millions of individual financial situations and decisions.
There will likely also be some excellent open-source, public-domain versions for people with a little more technical sophistication and desire for privacy. But the total fees collected will likely be a tiny sliver of today’s investment management fees.
The investment management business, moreover, is only one of many in which AI will cause radical revenue reductions for providers and massive gains for individual consumers.
Parting Thoughts
Things, of course, could work out quite differently. But risk managers should certainly consider scenarios in which trillions of dollars of asset management market cap evaporates in the medium term. From a career standpoint, risk managers — and all knowledge workers for that matter — should think about what core competencies they can bring to an AI world. If those seem weak, consider learning some new ones.
A fascinating theoretical question is whether a market dominated by AI traders can eliminate the booms and busts, bubbles and crashes, endemic to human-run financial markets for all previous history. Some disruptions are clearly caused by inattention, irrationality and frictions, things AI should be able to avoid. But perhaps some are fundamental to dynamic, innovative economies, and will persist, even with constant detailed attention and continuous rational adjustments. Either way, though, financial markets will be fundamentally different.
This won’t happen immediately, but I’d say in five years humans in investment management will be working for the AI, rather than getting AI to work for them. In 10 years, some early adopters will be experimenting with personal AI financial managers. In 20, a human making a financial decision will seem like an airline pilot turning off the GPS and autopilot to fly the plane by stick and dead reckoning.
Aaron Brown worked on Wall Street since the early 1980s as a trader, portfolio manager, head of mortgage securities and risk manager for several global financial institutions. Most recently he served for 10 years as chief risk officer of the large hedge fund AQR Capital Management. He was named the 2011 GARP Risk Manager of the Year. His books on risk management include The Poker Face of Wall Street, Red-Blooded Risk, Financial Risk Management for Dummies and A World of Chance (with Reuven and Gabriel Brenner). He currently teaches finance and mathematics as an adjunct and writes columns for Bloomberg.
This article was adapted from a commentary originally published on Bloomberg.com.
Topics: Risks & Risk Factors