When it comes to A.I., investors have a torturous choice to make: Believe the hype and jump in headfirst, or ignore the hype and potentially miss out on a once-in-a-generation tech transformation. “It’s the early days in terms of picking winners and losers” of the A.I. boom, explained Scott Helfstein, head of thematic solutions at Global X ETFs. “We’re not sure yet of the different ways that we’re going to be able to monetize this technology. Some are obvious, and some are not obvious—and it might be that the ones that are not obvious become more important over time,” he said. According to analysts at PricewaterhouseCoopers, A.I. could add up to $15.7 trillion in global GDP in 2030, which is more than the current output from India and China combined.
But experts caution that investing in generative A.I. is far more nuanced than just buying Big Tech stocks that have integrated the technology. “This could be another one of those moments where we’re getting to the peak of the hype cycle, so companies are going to spend the money, but they’re going to find it’s a lot harder to build something truly compelling,” explained Peter Cohan, business professor at Babson College. “Microsoft has a big investment in OpenAI, and they also have GitHub, which has a lot of the software that can be used for developing A.I., yet I still don’t know if that’s going to generate enough revenue for Microsoft for A.I. itself to become a significant part of its business,” he noted.
But it’s impossible to ignore that in an era of high inflation and subpar index returns, companies and funds that focus on A.I. and machine learning have strong growth potential. For example, the Global X Robotics and Artificial Intelligence ETF (BOTZ), which tracks 44 holdings of companies involved with industrial robotics and automation, nonindustrial robots, and autonomous vehicles, is up 18% year to date, while the S&P 500 is just up 7%. For investors who want to get in on this burgeoning area, here are the smartest ways to approach investing in A.I.
A plethora of companies across sectors are integrating A.I. into their operations, from Land O’Lakes to Microsoft to cybersecurity firm SentinelOne. So you may already be indirectly investing in A.I. Big Tech is leading the charge from both investment and innovation standpoints. “A.I. is being defined actively by leading tech giants,” explained PitchBook tech analyst Brendan Burke, citing the leaders as Microsoft (MSFT), Alphabet (GOOGL), Amazon (AMZN), and Meta (META). “Those companies have the research labs and the operational tools to bring basic innovations in A.I. algorithms to market,” he said.
If you want to bet on Big Tech, Microsoft has forged ahead the furthest with its $10 billion investment in OpenAI and the subsequent rollout of revamped search service Bing with new A.I.-powered search features. While Bing’s features have debuted with some bizarre and disturbing interactions with users, if the technology does successfully upend Google’s search, Microsoft has a huge market share to gain. “We believe incorporating ChatGPT capabilities into Bing may provide Microsoft with a once-a-decade opportunity to unseat Google’s Search dominance,” wrote D.A. Davidson software analyst Gil Luria in a recent research note. Luria increased Microsoft’s price target from $270 to $325 and gave the firm a buy rating. “While Google has announced a project to add [generative] A.I. to Google Search, we believe Bing’s head start could create a permanent share shift,” Luria wrote. Yet with the speed bumps in the rollout of Bing, it is clear that generative A.I. chatbots still have a long way to go.
Bank of America analysts, on the other hand, wrote in a recent report that Meta’s stock has the most to gain from A.I. implementation, citing the company’s incorporation of A.I. into its Reels video feature. “We expect an outsized impact from the adoption of A.I. given rich consumer data sets and massive ramp-up in A.I. investments, led by Alphabet, Meta, and Amazon,” the report detailed. Bank of America gave Alphabet and Amazon buy ratings in its forward-looking 2023 report.
Comparing the A.I. boom to the dotcom boom of the early 2000s, it is clear that if you pick the right companies you can still make out very well. “Even if you bought Amazon at the height of the dotcom boom, you still would have made a lot of money,” said Tom Taulli, a startup advisor and A.I. author.
But the giants have a problem—they’re huge. “It’s harder to move the needle for [Big Tech] as opposed to some of these smaller or midsize companies that are going to wind up playing in space that is going to be significant,” Helfstein said. “Maybe they won’t be as recognizable, but they have better growth potential, which can turn into investment returns,” explained Helfstein.
The “picks and shovels” of the A.I. boom
During the California gold rush, some of the savviest investors eschewed chasing gold itself and made a mint investing in the picks and shovels that those looking for riches needed for their search. Cohan sees an analogy in A.I. “It might well be that the initial beneficiaries of A.I. expansion are going to be in the hardware space, instead of the software space,” Helfstein predicted.
So what “picks” do experts recommend? PitchBook’s Burke explained: “A.I. growth is based on the improved efficiency and the decreased cost of [graphics] processing units (GPUs). The most basic enabler for these new models is the latest hardware innovations coming from chip giants such as Intel (INTC), Advanced Micro Devices (AMD), and Nvidia (NVDA).” Semiconductors are necessary for A.I. software to function, and the companies that produce them will supercharge its expansion.
Nvidia is currently the leading chip producer in the A.I. space. “A lot of capital is going to go into these generative A.I. startups,” said Cohan. “And the question for investors is: What technology will they be using to build their APIs? The answer is that they’re very likely to be using Nvidia chips.”
Beyond chips, there is software infrastructure to support the huge volume of data that the A.I. explosion will rely on. “The next layer up, there’s a need for enterprises to integrate their data into centralized data warehouses to improve the performance of their A.I. models,” Burke explained. “That’s where database management vendors, such as Snowflake (SNOW) and Alteryx (AYX) can help enterprises,” he said. Zacks gives both companies buy ratings.
Dynatrace (DT) is another cloud-computing infrastructure firm that could have upside. The company operates a software intelligence platform that uses A.I. and automation. Bank of America thinks the stock, which is up 1% year to date, has further to run, giving it a buy rating.
Another winner could be Palantir Technologies (PLTR), which is projected to increase its market share and revenue as one of its biggest clients, the U.S. government, expands its military A.I. implementation. In October 2022, the company signed an $85 million contract with the U.S. Army Materiel Command to use its software to increase military A.I. capabilities. The company was founded in 2003 by a group of VCs including Peter Thiel, Nathan Gettings, Joe Lonsdale, Stephen Cohen, and Alex Karp. While the stock is up 31% year to date, analysts (like those at BofA who have a “buy” rating on the stock) expect it can further capitalize on the A.I. boom.
Though A.I. technology has had a breakthrough year, there are already a robust number of funds and ETFs that give you exposure to the sector, which is a win for investors. “With these emerging technologies, people are better off playing a basket of companies, rather than putting their chips in a single one,” said Helfstein. He drew a comparison to Tesla in the electric vehicle industry. While Tesla was first out of the gate to dominate the EV market, competitors are now lining up to surpass the reigning EV king: “I don’t think we’re gonna live in a world that’s dominated by a single A.I.—it’s going to be a competitive market.” Not to mention the inherent unpredictability of an industry evolving at warp speed. “This can be a volatile space; you never know what could happen tomorrow, so diversification is important,” explained Taulli.
A solid option is the Global X Robotics and Artificial Intelligence ETF (BOTZ), which is up 18% year to date and has $1.64 billion in assets under management. Its top holdings include Nvidia, Intuitive Surgical, Upstart Holdings (UPST), Keyence (KYCCF), and ABB (ABB).
BlackRock’s iShares Robotics and Artificial Intelligence Multisector ETF (IRBO) is another option, which has 118 holdings and around $288 million in assets under management. Year to date, the fund is up 15%, solidly outpacing the S&P 500.
A widely diversified option is the ROBO Global Robotics and Automation Index ETF (ROBO), which holds 79 stocks with the top five holdings representing just 9% of the fund’s value. Year to date, the ETF is up 12%.
The Defiance Machine Learning and Quantum Computing ETF (QTUM) has roughly $116 million in assets under management, and its top holdings include Nvidia and Alteryx. Year to date, its return is roughly 16%, and the fund received a five-star rating from Morningstar Research in December 2022.
Overall, experts advise caution—and patience—when it comes to investing in generative A.I. “Investors should keep in mind that breakthroughs tend to be nonlinear,” explained Helfstein. Having just seen a period of what he calls “exponential innovation,” don’t be shocked if now we “level off for a little bit” before the next surge.
A version of this article appears in the April/May 2023 issue of Fortune with the headline, “Buying the A.I. boom.”