On March 14, 2023, artificial intelligence laboratory OpenAI released GPT-4, its latest large language model, or LLM.

Despite being released less than four months after its predecessor, GPT-3.5, the new model is capable of much greater feats of intelligence. OpenAI claims GPT-4 scores in the top 10% on the Uniform Bar Examination and the top 12% on the Law School Admission Test. For comparison, GPT-3.5 scored in the bottom 10% and bottom 40%, respectively.

The potential economic impact of this kind of AI advancement is hard to put a number on. In a 2022 report, accounting and consulting firm PwC, also known as PricewaterhouseCoopers, estimated that AI could contribute $15.7 trillion to the global economy by 2030.

Given these numbers, it’s no wonder that AI stocks have started to get attention from investors.

What are AI stocks?

Haydar Haba is the founder of Andra Capital, a venture capital firm that invests in AI companies. He said in an email interview that there are several publicly traded companies that have substantial AI interests and are poised to benefit from the growth of the industry.

AI stocks tend to fall into one of two categories: blue-chip technology companies that have invested in or partnered with AI developers, and small, experimental companies that are completely focused on AI development.

Shares of small AI developers might seem like the most “direct” investments in AI. But Michael Brenner, a research analyst who covers AI for FBB Capital Partners, says they’re not necessarily the best AI investments.

“Large language models require a tremendous amount of data and a huge amount of capital to put together,” Brenner says.

Brenner notes that small companies may develop innovative new models on their own, but eventually they will have to partner with a bigger company that has more infrastructure in order to run those models at a commercial scale.

“So far, we’re sticking with more of the mega-cap tech companies,” Brenner says, referring to FBB Capital Partners’ AI portfolio.

7 examples of AI stocks

Below is a list of AI stocks that Brenner or Haba highlighted as potentially interesting to investors:

• NVIDIA (NVDA). “NVIDIA today is a leader in AI and machine learning hardware, providing powerful [graphics processing units] for training and inference workloads,” Haba said.

• Taiwan Semiconductor Manufacturing (TSM). “All the chips powering AI have to be built by a very high-end semiconductor manufacturing facility, and Taiwan Semiconductor owns those,” Brenner says.

• Meta Platforms (META). “Meta has been invested heavily in AI, particularly LLMs,” Haba said.

• Amazon.com (AMZN). “The cloud companies, which already have lots of the chips and infrastructure that people need to run these large AI models, are definitely going to benefit from more demand for AI computing,” Brenner says.

• Microsoft (MSFT). “Microsoft has been investing heavily in AI research and development, with the $10B OpenAI investment being quite notable,” Haba said.

• Alphabet (GOOG). “Google has also been heavily investing in AI research and recently announced a strategic investment in Anthropic, a competitor to OpenAI’s foundation models,” Haba said.

• Snowflake (SNOW). Haba mentioned Snowflake as another cloud computing company that could benefit from increased demand for AI computing infrastructure, along with Microsoft, Amazon and Alphabet.

How to find AI ETFs

Haba said investors can also get exposure to AI stocks through exchange-traded funds that invest in a basket of companies involved in AI development and implementation.

There are several AI stock indexes — like the Indxx Global Robotics & Artificial Intelligence Thematic Index and the Nasdaq CTA Artificial Intelligence & Robotics Index — which are tracked by ETFs.

You can also find AI ETFs in an ETF screener by searching for “artificial intelligence,” “machine learning” or “AI,” but make sure to thoroughly research any ETF you find this way.

Some ETFs that have “AI” in their name invest in AI-linked companies. But others are diversified ETFs that use AI-powered trading, and are not necessarily invested in AI stocks.

Could we be in an AI bubble right now?

In the last few months, some small companies have seen rapid increases in their stock prices after being mentioned in any kind of AI-related news. BuzzFeed (BZFD), for example, saw its stock price soar more than 85% the day after the publication of a news report about a partnership with OpenAI to develop an AI-powered article writer.

Haba said that there are some signs investors may be overenthusiastic about the potential of AI right now.

“AI has been used as a buzzword to drive share price premiums, but companies have not always demonstrated usage of cutting-edge techniques,” he said.

“In the private markets, we are seeing hundreds of millions of dollars flow into ‘AI’ companies that have no product market fit. While investors are justifying such investments based on growth potential, it remains unclear whether that potential can or will ever be monetized,” Haba said.

Brenner also acknowledges that some individual AI-linked companies have seen their valuations increase sharply without a big change in their business fundamentals.

However, Brenner doesn’t think that the current surge of interest in AI amounts to a bubble, because many indexes of technology stocks are still below their 2021 highs.

“At macro-level, it’s hard to say we’re in a bubble, because we’re not at an all-time high,” Brenner says.

Should investors buy AI stocks?

Investors should think carefully before buying individual stocks or narrowly focused ETFs. Targeted investments can be riskier than broad investments like S&P 500 index funds, which have a long-term average annual return, not accounting for inflation, of about 10%.

One guideline that can help limit that extra risk is to devote no more than 10% of your overall portfolio to individual stocks.

But if you’re financially secure enough to buy individual stocks, Brenner says AI stocks are worth considering.

“I think retail investors should be thinking about how machine learning is going to impact the stocks they own,” he says.

“For individual investors looking at individual stocks in their 401(k)s, if they can weather the volatility of individual stocks, investing in AI-related companies as part of a diversified portfolio could make sense,” Brenner says.

The author owned shares of Alphabet at the time of publication.

LeackStat 2023