This is part one of a four-part series looking at the AI boom, the hype, the numbers, and what might come next.
I hear that markets are in a bubble, but also that AI is going to take our jobs. Two statements that contradict each other, yet both capture the mood of the moment.
Markets have soared on the back of AI. Valuations are pushing into record territory, and the headlines are full of talk about the “next big thing”. But beneath the frenzy lies a more grounded question: are we experiencing a speculative bubble, or is this something more enduring? The difference matters, because the answer shapes how investors behave.
Today’s constant fretting about whether frothy tech stocks are in a bubble is rooted in fears that AI may not live up to the hype, that business returns may never justify the blistering levels of investment being poured into advanced chips, infrastructure, and the energy needed to power AI’s vast processing demands. Columns will be written on the “AI hype” question in the months ahead, most likely concluding that it’s too early to tell. Companies are not yet reaping significant returns from AI, even as an estimated $3 trillion of investment is expected to be ploughed into the technology over the coming years. But generating returns was always going to take time.
Let’s step back and compare where we are now with where we were in the dot-com era, and then ask whether the numbers actually support today’s excitement.
Valuations Are High — Historically High
Make no mistake, valuation multiples are elevated. In the tech sector today, forward price-to-earnings ratios are well above long-term averages. Tech P/Es have moved from around 20 times earnings at the start of the year to closer to 24 by October, comfortably above both five and ten-year averages. At the same time, many private-market AI firms are commanding large valuations despite still being in the early stages of revenue growth.
On the surface, that looks a lot like bubble behaviour, a surge of optimism, a flood of capital chasing ideas, and a growing gap between price and business fundamentals. Year-to-date, AI and cloud leaders have far outpaced the broader market, feeding the sense that tech valuations may be running hot.
What’s Different This Time
When you look back to the dot-com bubble of the late 1990s and early 2000s, you see a striking contrast, high valuations paired with weak profits. Many companies were valued purely on promise, with little or no earnings to show for it. Today the story is different. Many of the leading tech firms tied to AI aren’t just promising, they’re earning.
For instance, the proportion of unprofitable companies in the tech sector has fallen from around 36% in early 2000 to roughly 21% earlier this year. In other words, valuations are high, yes, but so is profitability. Valuations alone don’t automatically make a bubble.
How Profits Are Holding Up
The profit angle is critical. If valuations are high and profits are low, that’s classic bubble territory. But this time, many companies are posting healthy margins. Nvidia, for example, reported net margins of just over 50% in its Q2 2025 results, an extraordinary level for such a large company.
It’s also worth noting that major tech players are already integrating AI in ways that drive both revenue and efficiency, rather than just future promise. Valuations may be near all-time highs, but so are net profits. That’s a big difference. It shifts the dynamic from hope meets hype to hope plus earnings.
Why “Bubble” Still Remains on the Table
Still, profits alone don’t mean we’re out of danger. There are genuine risks investors shouldn’t ignore. A large share of the gains are coming from a handful of dominant AI winners, which makes the market vulnerable if even one stumbles. Much of the current AI revenue remains experimental, pilot projects, proofs of concept, or early stage services that may not scale as easily as expected.
Building AI infrastructure is also capital intensive. Data centres, compute power and energy networks require vast investment. If those costs rise faster than returns, the support for high valuations could quickly weaken. And perhaps most importantly, expectations are already sky-high. With long-term growth largely priced in, there is very little room for disappointment. Even a small miss in execution could trigger a significant correction.
The recent pullback in tech shares seems to underline that point. Reports suggest the sell-off was largely positioning-driven, with the year’s biggest outperformers hit hardest. There was no clear trigger, in fact, it began after a round of stronger than expected financial results. That suggests short-term profit taking rather than panic, as fund managers, conscious of year-end performance, step aside briefly rather than heading for the exit.
The Bottom Line
Put it all together and the picture is nuanced. Yes, valuations are rich. Yes, there’s some froth at the edges. But this does not appear to be a repeat of the dot-com bubble. The difference is meaningful: strong profits, real adoption and widespread corporate commitment.
The companies that combine sound valuations, sustainable earnings and credible business models tend to stand apart from those built purely on excitement. High valuations make execution matter.
Markets are, as always, driven by perception. Right now, that perception is that AI is transformative. Whether results live up to that belief is what comes next.
Yours
SW
This article is for informational purposes only and should not be taken as financial advice. SmartWealthHQ aims to share insights and education to help readers make informed choices. Always do your own research.