As the US stock market reached record highs this October, investors have questioned whether they are heading for another tech bubble, driven by enthusiasm and hype about artificial intelligence.
While the market shows signs of getting bubbly, some strategists would argue it is still nowhere near a point where it’s in a bubble that is ready to burst, as it was in the year 2000 when the Internet drove a tech stock craze. Now, the biggest US technology companies have become part of the AI boom.
“You’ve got probably the biggest generational shift in technology since the Internet, and you’re only three years into this,” said Dan Niles, founder and portfolio manager with Niles Investment Management. “Netscape Navigator, which was arguably the first mass Internet web browser, came out in 1994. It took over five years for that bubble to reach its peak. We’re only three years into the launch of ChatGPT since the end of 2022. From a time perspective, it takes a certain amount of time for these bubbles to inflate.”
Niles expects the tech rally to continue, with AI developments like video generation apps, bringing new interest. “Why can’t this go on for another year?” he asks.
The broad tech sector is also powered by real earnings growth. In the late 1990s, there were many Internet-related stocks with high valuations and no profits. The S&P information technology sector is on track to realise a 24.8 per cent earnings gain for the third quarter, according to CFRA Research. Earnings for the sector are forecast to be up 20.8 per cent for 2025 and about the same for 2026.
“With valuations being as stretched as they are, I think we need a continued stream of better-than-expected results to justify them,” said Sam Stovall, chief investment strategist with CFRA Research.
A number of the biggest tech companies reported earnings this past week and did better than expected. But announcements of higher spending plans sent stocks of two tech giants lower. “Even though Meta and Microsoft were punished after their earnings reports, we still have to buy ratings on the stocks, based on longer-term growth potential,” said Stovall.
Alphabet, Amazon and Apple also released earnings, and their stocks gained immediately following their reports.
“We’re not surprised that the latest leg of the secular bull market in assets, tech and US tech in particular, is being led by the AI revolutionary companies mostly centred in the US,” said Julian Emanuel, Evercore ISI head of equities, derivatives and quantitative strategy. Emanuel, who had expected the major tech companies earnings to beat expectations, believed “the fundamental commentary around earnings completely ratifies the long term trend”.
However, he noted that while the price action and investor bullishness signal a potential pullback, the longer-term rally should continue. “Ultimately, these stocks will become overvalued,” he said. “But we still think you’re a long way from there.” He expects the S&P 500, which finished at 6,822 on Thursday, to reach 7,750 by the end of 2026.
“The names and the themes that have been leading are likely to be the ones that continue to lead to 7,750,” he said.
However, as Big Tech continues to spend on AI, investors will continue to watch capital expenditures closely. “There’s a lot of hand-wringing about whether this is another bubble. It may get there at some point in time,” said Art Hogan, chief market strategist at B. Riley Wealth. “It certainly isn’t there now. The 10 largest AI players are companies with business models. They throw off massive revenues, can actually pay for this capex and are growing earnings at an incredible rate.”
Bank of America said in a recent report that capex spending by the large US hyperscalers could reach $1.2 trillion over five years. Those companies include Amazon, Alphabet, Microsoft, Meta and Oracle. “My larger concern remains that by late 2026, AI revenues will not be ramping enough to validate the enormous ramp in spending in AI capex,” said Niles.
He added that while the big hyperscalers are expected to spend more than $800 billion from 2023 through the end of this year, “the revenues generated by the AI native companies such as OpenAI, Anthropic and Perplexity are only expected to reach about $20 billion this year”.
Ultimately consumers will have to pay more for AI. “The ultimate question is you’ve got to get something to pay for this, and consumers have been trained that it’s free,” said Niles.
Tom Lee, founder and head of research at Fundstrat, remembers how the Internet bubble formed and continued for several years, but the current market is not nearly as overvalued.
While the price-to-earnings ratio for some tech companies is very high, it does not compare to the valuations in the year 2000. The price-to-earnings ratio or P/E is just what it says it is – the value of the company’s expected earnings relative to its stock price. For instance, the S&P 500 P/E ratio is about 24.1 times based on the amount of earnings expected to be generated by the S&P 500 companies in the next 12 months.
Nvidia’s forward P/E is currently about 35, while Lee notes that Cisco, the hottest stock of the tech bubble, has a multiple of over 200. Apple’s forward P/E is about 34.7. “I don’t think it’s at bubble territory. You know you’re not in a bubble when everyone says it’s a bubble,” said Lee.
CFRA’s Stovall warns that any hiccup could trigger a stock market sell-off between now and the end of the year, and that could ultimately provide a buying opportunity.
“Right now, the S&P 500 is trading at a 43 per cent premium to its 20-year average forward P/E, and tech is trading at a 72 per cent premium. A pullback is a possibility,” he said. He likens the P/E to square feet in real estate construction. The higher the price per square foot, the pricier the property — or in this case, a stock.
“We still like tech, but we do for the long term,” said Stovall. He said historically when there’s a new decline in the S&P 500 of five per cent or more in the same calendar year that started with a deep correction, the average decline has never been more than nine per cent “That’s encouraging, and it means we’re not heading for a bear market or even a correction. It would be more of a pullback,” he said.
Niles said investors who want to diversify US tech holdings may want to consider Chinese tech. Startup DeepSeek’s release of its AI model earlier this year surprised investors and put a spotlight on the Chinese industry.
Niles said the Chinese government has become more supportive of its companies. Its tech industry also does not face the same energy issues as the US, which needs to build out more electric generation to power the growing AI industry. “US tech is the world leader, but China must definitely be considered,” he said. “China’s in a pretty good spot, and the valuations are nowhere near as extended as they are in the US for a lot of these names.”
Investors in US technology should also realise that when they buy the S&P 500 index, it is heavily influenced by the Big Tech stocks.
Buying individual shares of Big Tech companies is one way to trade them. There are also mutual funds and ETFs that focus on different themes and sub-sectors. Analysts caution that investors should also know what holdings they have in mutual funds and ETFs since they may be taking a bigger tech stake than they expect.
Strategists expect there will be market sell-offs, but the AI investing theme should continue to be a dominant one. “Was the Internet overhyped in the late 1990s? Yes, but was it underhyped long-term? Yes. I think it’s the same thing with AI,” said Niles.