We are all investors now, or at least the majority of us are. More Americans than ever are investing in the stock market, and surveys show a large share of Gen Z and millennials actively trade individual stocks as their preferred investment strategy. Most people lose money or underperform with this latter endeavor. But take a bull run and technology that makes trading feel like a video game, and suddenly, everyone thinks they can beat the market.
I tell myself I am just glad people are so engaged with the economy. But now, we are taking it to another level: Agentic AI promises to help everyone beat the market and get rich from their basement — though under-performing the market is probably the best individual investors can hope for.
Robinhood Markets Inc., the fastest-growing trading platform in the US, last week announced a new suite of features, including greater access to its agentic AI trading tools. “We’re making Robinhood the best place in the world for active traders by delivering tools once reserved for hedge funds, big banks and quant firms,” Chief Executive Officer Vlad Tenev said at the launch.
See more: AI is Taking Over the Value Style Too
Everything about his comment is revealing. The name Robinhood suggests that it is taking money from the rich (hedge funds, family offices, etc.) and giving it to the poor (some guy in his parents’ basement). Tenev also implies that making money in markets just takes some advantage that retail investors are usually deprived of. That advantage now is AI. Rival trading platforms are sure to release competing products.
Robinhood users could already integrate their own AI agents on the platform. But the company wants customers to use its AI-powered agents that can take in information and automatically trade on all sorts of assets, including exotic products, 24 hours a day, 7 days a week. What could go wrong?
The first worry is that the technology will hallucinate data and make a bad trade. Or maybe it will go rogue and bankrupt you. Or maybe the same technology using the same data for so many people will increase herding and, therefore, systemic risk.
Recall the flash crash of 2010 caused by a high-frequency-trading algorithm. AI may be smarter, but that experience shows the dangers of relying on automated trading. Tenev says the default setting is for a human to approve every trade. But he concedes that customers who want speed can turn that feature off, and he expects many people will opt to do so.
Even if you don’t have these concerns, odds are you will not make more money with an AI trading bot — even a good one.
Every active trader has the firmly held conviction that they are either smarter than everyone else or have better information. Maybe you notice a promising company in your experience as a consumer, maybe you get a great (but legal) tip on a Reddit thread, see a pattern in the data no one else does, gain a special insight from a footnote in an earnings report, or are just faster than the next guy. If you believe any of this is how you make money, it stands to reason AI can beat the market; it is better at seeing patterns and scanning earnings reports, and if you empower it to do so, it can trade that information instantaneously.
But for most individual investors, this is not how things work. First, active trading is zero-sum and most people work off the same information that gets priced in very quickly. There is no secret magic pattern hiding in the data. Even if there were, with enough people using AI, that magic pattern would rapidly disappear. It is not clear how that translates into durable alpha for a retail trader.
Any mispricing that can be exploited to make money is likely small, because markets are already very efficient at pricing in information. Earning a big profit then requires amplifying it with lots of leverage. This does give the big hedge funds and institutional investors an advantage because they have more money and access to credit. AI-powered Robinhood can’t overcome that reality, nor should it, since it is a very risky strategy.
AI has not proven itself to be better than the average trader, so far. Like humans, it tends to be bad at market timing. It could get better, but the most we can expect then is for information to be priced in even faster. That would make markets even more efficient and enhance liquidity and capital allocation — all beneficial for the economy but not a retail investor’s wallet.
Agentic trading is just the next step in an evolution that brings us closer to the point where trading employs less human judgment but, presumably, prices reflect information better.
When high-frequency trading took over markets in the 2010s, it made it much harder for some traditional active managers to make money because the algorithms were faster and better at spotting anomalies. It also lowered spreads and increased efficiency.
As long as I’ve studied finance, I’ve been told that markets are inefficient because they reflect all our cognitive biases; that our so-called irrationality is why pricing anomalies exist, and how clever, more rational individuals beat the market. With the eventual takeover of agentic trading, which is supposedly rational and data-based, what’s an active trader’s reasoning now?
This leaves a retail Robinhood investor with two options. Be part of the price discovery process, and risk losing money in a zero-sum game where you have no advantage (and that assumes your agent doesn’t go rogue). Or enjoy the fruits of everyone else’s losses and buy an index fund.
A message from Advisor Perspectives and VettaFi: Discover something new! Click here to register for our upcoming webcasts.
Bloomberg News provided this article. For more articles like this please visit
bloomberg.com.
Read more articles by Allison Schrager
