As ChatGPT approaches its third anniversary, a surprising trend has emerged: at least one in 10 retail investors is now using a chatbot to select stocks. This has fueled a boom in the robo-advisory market, but even its supporters admit it’s a high-risk strategy that can’t fully replace traditional financial advisors just yet.
Accessing Investment Insights for Everyone Thanks to artificial intelligence, anyone can now select stocks, track their performance, and get investment analysis that was once exclusive to major banks and institutional investors. The robo-advisory market, which includes all automated, algorithm-driven financial advice providers like fintech companies, banks, and wealth managers, is projected to grow from $61.75 billion last year to $470.91 billion by 2029. This represents a massive increase of roughly 600%, according to data analysis firm Research and Markets.
The Limitations of AI Jeremy Leung, who spent nearly two decades analyzing companies for UBS, has been using ChatGPT to pick stocks since leaving the Swiss bank last year. “I no longer have the luxury of a Bloomberg terminal, or those kinds of market-data services which are very, very expensive,” Leung said. He noted that even a basic ChatGPT tool can replicate many of the workflows he used to perform. However, he cautioned that these tools might miss key analysis because they cannot access data hidden behind a paywall.
Leung is not alone. The industry is expanding at a rapid pace. A survey by broker eToro found that about half of all retail investors would use AI tools like ChatGPT or Google’s Gemini to make or change investments in their portfolios, and 13% are already doing so. In the UK, a survey by Finder revealed that 40% of respondents have used chatbots and AI for personal financial advice.
“Use Only Credible Sources” Even ChatGPT itself warns users that it should not be relied on for professional financial advice. Dan Moczulski, UK managing director at eToro, highlights the danger, stating, “The risk comes when people treat generic models like ChatGPT or Gemini as crystal balls.” He recommends using AI platforms specifically trained to analyze markets, as general models can misquote figures, rely too heavily on past trends, and fail to predict future market shifts.
In a test, Finder asked ChatGPT in March 2023 to select a basket of stocks from high-quality businesses. The resulting selection of 38 stocks, which included Nvidia and Amazon, as well as consumer staples like Procter & Gamble and Walmart, has since surged by nearly 55%. This is almost 19 percentage points higher than the average performance of the UK’s 10 most popular funds.
Leung’s experience shows that getting the best results requires a level of financial knowledge. He creates prompts like “assume you’re a short analyst, what is the short thesis for this stock?” or “use only credible sources, such as SEC filings.” He believes that “the more context you provide, the better the responses.”
The risks are significant. The excitement over this new tool, which has democratized access to investment information, also makes it impossible to know if retail investors are properly using risk management tools to mitigate potential losses when the markets inevitably turn. “If people get comfortable investing using AI and they’re making money, they may not be able to manage in a crisis or downturn,” Leung said.
