Three AI stocks to watch

Artificial intelligence (AI) stocks are still all the rage and look set to be for some time into the future, even despite growing concerns over their valuations.

The top stocks in the market are heavily saturated, and talk of the AI megacap bubble bursting is gathering momentum. The volatility in this space is captured neatly by Nvidia’s share price falling over 7% during the first week of November as concerns over AI valuations have mounted.

The week “ended very much on the back foot, as AI-focused US tech stocks lost close to $1 trillion in value,” said Tom Stevenson, investment director at Fidelity International.

MoneyWeek

Subscribe to MoneyWeek today and get your first six magazine issues absolutely FREE

Get 6 issues free

Sign up to Money Morning

Don't miss the latest investment and personal finances news, market analysis, plus money-saving tips with our free twice-daily newsletter

Don't miss the latest investment and personal finances news, market analysis, plus money-saving tips with our free twice-daily newsletter

Sign up

But Stevenson played down talk of the AI boom having run its course. “Talk of another bubble still looks wide of the mark,” he said. “Between 1998 and 2000 the valuation of the 50 biggest stocks in the S&P 500 doubled from 20 to 40 while those of the other 450 stocks remained below 20. This time around the gap between the two is much narrower – 29 versus 24.”

Even if it does turn out to be the ultimate bubble, there are still avenues for investors to find value in AI.

These picks of AI stocks to watch all have something that links them to the AI theme in the bull case, and could see their performance improve significantly should the AI boom continue to accelerate as its proponents expect. But equally, they have other strings to their bow (and so aren’t entirely dependent on AI), and/or are trading at valuations that make them less risky than the headline names in the event of the AI bubble bursting.

Read on to discover three interesting AI stocks to watch as we come towards the end of 2025.

Qualcomm’s inference bid

The semiconductor stock par excellence in my opinion is Taiwan Semiconductor. These days, though, it’s less of a hidden gem than it used to be, though its forward P/E multiple of 25 is still attractive given its near-monopoly on supplying AI giants trading far above there.

So while Taiwan Semi will always be worthy of a special mention, right now one of the more interesting semiconductor stocks to watch is Qualcomm (NASDAQ:QCOM). Until recently it was happy to sit on the sidelines as far as AI was concerned, and concentrate on chips for wireless connectivity and mobile devices.

But on 27 October Qualcomm announced it was throwing its hat into the AI inference ring, with two chips, the AI200 and AI250, expected to launch in 2026 and 2027 respectively.

Inference refers to the process where AI models digest prompts and generate responses, as opposed to training, which is the stage where model developers like OpenAI effectively teach them how to operate. Inference is a more data-intensive process than training and as such could be a lucrative market for chipmakers.

The market responded positively to the news – Qualcomm shares gained over 11% on the day it was announced – but they have drifted back down since.

Going up against the likes of Nvidia and AMD in inference chips will be no walk in the park for Qualcomm, but with shares trading at just 14 times forward earnings (compared to around 30 times for Nvidia and AMD), expectations are far more measured, and its strong position in wireless chips further mitigate the downside risk.

Following its latest results, Bank of America analysts raised their price objective from $200 to $215 (26% above the close price on 7 November).

Alphabet’s gathering cloud

Alongside chips, cloud services is a fundamental component of the AI infrastructure which, unlike (for now) deploying large language models (LLMs) themselves, is clearly highly profitable.

Three cloud service providers control 62% of the market (based on Q3 data from Synergy Research): Amazon (with 29%), Microsoft (20%) and Alphabet’s (NASDAQ:GOOGL) Google Cloud (13%).

Google Cloud revenue increased 34% during Q3 2025, compared to 33% for Microsoft’s Azure and 20% for Amazon Web Services (AWS).

Morgan Stanley analyst Brian Nowak thinks Google Cloud could grow at up to 50% in 2026, well ahead of current Wall Street analysts’ consensus of 31%. That could see it bear down on Microsoft and even Amazon’s market share over coming years if it can maintain that growth rate.

Alphabet is also one of the most diversified companies out there. It has its own LLM (Gemini) and has developed its own custom chips for AI – and that doesn’t even touch on its core Search business which avoided a forced break-up in an antitrust case in September.

Despite all this, the market seems to heavily discount Alphabet compared to the other Magnificent Seven stocks; a forward P/E ratio of 23 seems very reasonable.

NRG Energy for the power play

One of the headwinds that the AI boom has to contend with is how energy hungry LLMs and data centres are. In the West especially, the price and availability of energy will likely be the key bottlenecks to the advancement of AI models.

So an obvious play is for companies providing that energy. Diversified suppliers like NRG Energy (NYSE:NRG) are seeing benefits from this already: NRG’’s earnings per share (EPS) increased 32% year-on-year in its most recent quarter.

The stock has almost doubled so far this year but still trades at around 20 times forward earnings, with analysts polled by London Stock Exchange Group expecting annual EPS to increase by over 40% between 2024 and 2026.

The author holds shares in Alphabet.