Can AI Tools for Picking Stocks Help Investors Beat the Market?

Most investors can’t beat the market, but that doesn’t stop them from trying. Humans are driven by fear and greed, which can lead to impulsive and often subpar investment decisions. So it’s no surprise that some are now turning to artificial intelligence for help.

The latest use case for AI, which is being marketed to both retail and institutional investors by companies like Danelfin and Boosted.ai, embraces the technology to sift through heaps of data in order to pick stocks so humans don’t have to.

Some of these companies claim their AI-powered platforms are capable of outperforming the S&P 500, which is an impressive feat given novice and professional traders’ inability to do so. However, results have been mixed so far, and the risks of using AI to help pick stocks are often similar to the risks inherent in other short-term or speculative investment strategies.

Even stock pros rarely beat the market

Because the average person isn’t equipped with an understanding of financial statements, technical analysis or how macrotrends affect sector performance, picking stocks can be a risky endeavor. That’s why Warren Buffett advocates for passive investors to buy shares of index funds like the SPDR S&P 500 ETF Trust, which has averaged a 10.54% annual return over the past 30 years.

Failing to beat the market isn’t specific to retail investors, though. A 2020 study conducted by S&P Dow Jones Indices compared actively managed funds to the performance of the S&P 500, finding that 89% of fund managers failed to beat the benchmark index.

More recently, BNP Paribas found that in 2023, hedge funds returned an average of 6.67% as the S&P 500 returned 24%. This, to some extent, explains the resurgent popularity of index funds among retail investors and hedge fund managers alike.

And while set-it-and-forget-it alternatives like robo-advisors have demonstrated some success, even their algorithms cannot compete with the average annual returns of the S&P 500.

However, for investors who remain captivated by the possibility of outsized gains but don’t want to choose their own stocks, there’s a new kid on the fintech block that’s garnering some attention.

Is AI good at picking stocks?

The AI technology being used for picking stocks isn’t your run-of-the-mill generative AI platform. ChatGPT and Gemini, which are large language models, are incredibly powerful tools but are limited in what financial advice they dole out. Inquiring about stock picks on those sites produces responses like, “I can’t recommend specific stocks, but I can help you with resources to do your own research,” and “As an AI language model, I can’t offer personalized investment advice.”

On the other hand, AI stock-picking platforms are specifically designed to do just that: choose and rate stocks and ETFs for humans. Considering how they remove time-consuming and jargon-laden research and analysis from the equation, it’s easy to see why everyday investors and professional traders alike are using the technology in an effort to beat the market.

That could be a stretch according to Andrew J. Evans, CEO and founder of Rossby Financial, a platform that provides support on technology, regulation and resources to help financial advisors.

“It’s a little silly to claim they’ll always beat the S&P 500,” says Evans. The AI “can’t tell the future. It has better predictive abilities, but individuals have to determine ‘Is this best for me?’”

According to Evans, these AI-driven stock-pickling platforms’ successes or failures will be determined by how well they perform during economic downturns. “If the market is raging, everything works,” he says. “Where [AI] will prove its mettle is the next bear market.”

The S&P 500 has firmly been in a bull run since October 2023, and it could be some time until the next bear market arrives. But when it does, if these AI stock-picking platforms outperform, that could help solidify their case as useful tools. Evans believes that in “three to four years, it’ll be a very solid predictive technology.”

One company aiming to change retail investing with AI

Danelfin, a stock analytics platform powered by AI, aims to improve investors’ chances of securing large returns by leveraging artificial intelligence’s ability to sift through massive amounts of data and provide retail investors with technology formerly only available to professionally managed funds.

Founded in 2016 in Barcelona, Danelfin's mission is to “democratize the use of artificial intelligence to help everyone make better investment decisions.” The company aims to do so by using its Explainable Artificial Intelligence, an analytics platform, to provide users with stock and ETF ratings and an easy-to-understand AI-generated score that ranges from 1 to 10.

It offers three plans. The free option includes a daily and monthly newsletter with top 10 AI-generated stock and ETF picks. Danelfin’s Plus plan ($19/month) provides unlimited picks, reports, rankings and alpha signals that can be used to make trades, and its Pro plan ($52/month) includes all of the above plus trading parameters with buy track records as well as the ability to export data as a CSV file.

The platform uses 600 technical indicators, 150 fundamental indicators and 150 sentiment indicators for every stock and ETF it rates. According to Danelfin, the higher the score assigned by its AI, the higher the probability that an equity will outperform the market over the next three months.

Does Danelfin work?

While companies like Danelfin promise to improve returns for investors, users shouldn’t rely solely on the technology to make investment decisions, but rather use it as a tool to better inform their choices.

“The onus is on the [AI] company,” says Evans. “Have they taught their machine to distill the information properly?” The next question the individual investor should ask themselves is: “Do I understand how this machine delivered this result, and am I comfortable with that result?”

To demonstrate its capabilities, Danelfin backtested its predictive AI vis-à-vis the S&P 500 from Jan. 3, 2017, to Aug. 15, 2023. In its modeling, the company’s AI-powered stock strategy generated a return of 191% versus the S&P 500’s 118% over the same period.

However, the company’s AI-generated recommendations are hardly guaranteed. Take, for example, its price targets for 3M, a $51 billion market cap conglomerate. Between January 2023 and April 2023, the company’s AI flashed dozens of “Buy” signals for 3M, resulting in some lofty expectations.

On April 5, 2023, when shares of 3M were trading for around $104, the platform generated price targets ranging from $145 to $208, which it expected the stock to hit by July 25.

In reality, 3M shares reached a high of just under $110 as of July 25. That represented a respectable 5% gain over the previous three months, but it failed to match the lowest price target generated by Danelfin’s platform.

The platform aims to project actionable trades within three-month windows. Buy-and-hold investors who purchased 3M as suggested on April 5, 2023, would be down nearly -11% one year later.

It's also important to keep in mind that while these services offer a form of assistive technology, AI-powered stock-picking platforms still require human decision-making. Evans, who is a big believer in machine learning, contends that it’s ultimately the investor — not the AI — who dictates what is or isn’t relevant to them.

“Time and time again, it’s going to make them appreciate the process more,” he says. “You can teach [predictive AI], but you still need a human to make the decision.”

A burgeoning tool for institutional investors

Another company, Boosted.ai, is using artificial intelligence specifically to assist professional traders. Co-founded in 2017 by Joshua Pantony, the former principal machine learning engineer at Bloomberg, and Nicholas Abe, a certified financial analyst, Boosted.ai holds 11 patents in the AI and fintech spaces.

Since its inception, the company has helped dozens of investment managers — whose assets under management exceed $1 trillion — implement machine learning in their portfolios. Its software, Boosted Insights, relies on AI to help institutional investors make more informed decisions.

According to the company, Boosted Insights “helps asset managers energize their equity portfolios with artificial intelligence….to find ways to slash hours from their research process, improve portfolio metrics, and have a holistic view of what affects their portfolios.”

While public information about Boosted.ai’s platform is limited, the company’s website shows it rated Waste Connections as a “Strong Buy” on Nov. 29, 2023. Since then, shares have risen an impressive 27%. However, the company’s AI-generated stock picks are only available to financial advisors and institutional investors.

But Evans believes it’s only a matter of time until these advanced tools are widely available to the general public. “Once big banks begin putting their data into a machine … give it two to three more years and we’ll have it.”

Are AI’s stock picks trustworthy?

The use of AI for picking stocks is still in its infancy, but it is rapidly evolving. While the technology may be more sound compared to relying on social media stock tips, for example, AI-assisted investing tools have thus far shown mixed results and appear best suited for experienced and professional traders. And although this technology can be used by novices, it’s unlikely to provide the desired long-term results by itself.

Simply put, Evans says of AI: “It’s not going to turn the general public into Warren Buffett.”

More from Money:

Why April Could See the Stock Market Move Even Higher

Best Stock Trading Apps of 2024

Why Hedge Funds Are Embracing Warren Buffett’s Classic Advice About Index Funds