Here are the 9 smartest ways to invest in artificial intelligence right now, according to Bank of America, Morgan Stanley, and UBS

Some on Wall Street are concerned that investors are becoming swept up in the hype surrounding artificial intelligence, but three top investment firms are increasingly confident that AI will live up to its lofty expectations.

Bank of America, Morgan Stanley, and UBS are more bullish on AI stocks after expressing reservations earlier this year as shares of companies associated with the technology skyrocketed. The companies explained why they’re now more comfortable with AI in separate late-July notes.

“We don’t think AI is a bubble given clear use cases and solid long-term visibility,” wrote Solita Marcelli, chief investment officer Americas at UBS’ global wealth management arm, in a July 25 note.

That vote of confidence came exactly two months after Marcelli’s colleague Art Cashin compared the rise of artificial intelligence to a “mini-version” of the dot-com bubble.

However, UBS now expects global AI demand to more than tenfold, from $28 billion in 2022 to $300 billion in 2027, according to Marcelli. Initially forecasting a 20% compound annual growth rate for AI demand between 2020 and 2025, UBS now forecasts a 61% compound annual growth rate from 2022 to 2027.

Bank of America is even more bullish on AI, predicting in March that the technology will be worth $900 billion globally by 2026. But, as with UBS, not everyone at the firm is convinced. Michael Hartnett, Bank of America’s investment-strategy chief, said in May that AI was in a “baby bubble,” though much of his concern was about interest-rate hikes that should end soon.

Morgan Stanley, on the other hand, has dismissed concerns that AI stocks are in a bubble.

“Inevitably, the market will compare AI to the dot-com boom,” Morgan Stanley analyst Shawn Kim wrote in a July report. “However, today’s leading AI companies are well established, with good cash flow characteristics (for the most part), unlike many of the smaller companies that were wiped out during the dot-com collapse, which lacked a viable business case.”

Instead, the firm’s concern about AI was related to its immediate impact on earnings, particularly if economic growth slowed. While AI investment is increasing, according to a Morgan Stanley poll, only 4% of chief information officers are spending heavily on new generative-AI projects.

“While we share that excitement, we believe it is premature to extrapolate such benefits across the entire economy and company earnings for this year, especially given we are still facing growing risk of cyclical headwinds, as data suggest the business cycle is increasingly at risk of a slowdown,” Morgan Stanley’s US equity-strategy chief, Mike Wilson, wrote in a July 24 note.

9 ways to invest in AI right now

Investors who believe AI is more than a passing fad should consider nine investments recommended by Bank of America, Morgan Stanley, and UBS in recent notes.

According to Bank of America, six semiconductor-related companies are expected to benefit the most from AI: Nvidia (NVDA), Marvell Technology (MRVL), Broadcom (AVGO), Synopsys (SNPS), Cadence Design Systems (CDNS), and Camtek (CAMT).

However, the firm stated that increased investments in AI may simply offset declines in traditional computing spending. According to Bank of America, capital expenditure on cloud computing is expected to fall below 10% per year through 2024. To that point, Alphabet and Meta Platforms both reported lower-than-expected capital expenditures in the previous quarter.

“We continue to see benefits for semis as AI investment ramps, but it will become prudent for hyperscalers to show profitability from AI projects, potentially leading to cautious data center capex spending patterns,” research analysts led by Vivek Arya wrote in a note Thursday.

Nvidia, the top-performing stock in the S&P 500 this year, is the clear standout among Bank of America’s AI stock picks. The California chipmaker is at the forefront of the boom, collaborating with companies like ServiceNow, Accenture, Snowflake, and Microsoft.

Morgan Stanley cited Nvidia and Microsoft (MSFT) as “clear winners” of the AI revolution. The software behemoth was an early investor in OpenAI, a pioneer in the field thanks to groundbreaking tools such as ChatGPT and the image generator DALL-E. Microsoft is now leveraging that relationship to incorporate AI features into software products like Microsoft Office.

In terms of chipmakers, Morgan Stanley agreed with Bank of America that the road ahead was divided. While demand for AI chips is high, it is only partially compensating for a drop in non-AI chip sales at Taiwan Semiconductor Manufacturing Co., according to Wilson.

AI and automation may result in significant labor cost savings for consumer-focused firms, but Wilson believes it is too early to buy companies solely for that reason. Similarly, enterprise AI adoption is in its early stages and may not result in revenue until next year.

Following this year’s massive rally, UBS also advises taking a cautious approach to AI investing.

“Given the segment’s rich valuations, we are waiting for a pullback to turn positive on the segment again,” Marcelli wrote.

However, the firm sees opportunities in midcyclical companies, which include internet firms and those in the software industry but exclude hardware manufacturers such as semiconductor firms.

“We expect a broadening of the AI market from the semiconductor and hardware-centric infrastructure layer to the software and internet-based applications and data models layer,” Marcelli wrote.

While AI infrastructure stocks have already been priced for strong growth, UBS believes software firms have an enticing risk-reward setup as AI demand soars and integration into everyday products continues. In fact, UBS predicted that the AI applications and data models segment would be worth $170 billion in 2027, compared to $130 billion for the hardware-infrastructure segment. To catch up, the industry will need to grow at a rate that is more than three times the annual rate, boosting stocks that are linked to it.