Jasmine Birtles
Your money-making expert. Financial journalist, TV and radio personality.
What’s Going On With AI Stocks in September 2026?
Moneymagpie Team
22nd Sep 2026
If you’ve felt a bit confused watching AI and chip stocks this month, you’re not imagining it. In the space of about three weeks, the AI trade has gone from “is this a bubble about to pop?” to a genuine sell-off to a fresh record high. All while some of the most powerful people in tech have been publicly arguing with each other about how fast AI should even be developed.
Here’s a recap of what’s actually happened this September, why the swings have been so sharp, and what it means if you’re a UK beginner investor with money in global funds, a pension, or individual tech shares.
The Short Version
- Early September: analysts and economists started warning that AI stocks looked like a “late-stage bubble,” pointing to stretched valuations and wild single-day swings in big tech names.
- 13-14 September: Anthropic CEO Dario Amodei published an essay calling for the AI industry to slow down its pace of development. OpenAI’s Sam Altman and Tesla’s Elon Musk publicly backed the idea.
- 15-16 September: the US Federal Reserve raised interest rates by a quarter point to 3.75%-4%, adding pressure on growth and tech stocks, while chip and AI stocks sold off sharply on the Amodei news, with some names down as much as 6% in a day.
- 16 September onwards: chip stocks including AMD began clawing back losses on strong company guidance and analyst support.
- 21 September: the Nasdaq hit a fresh record high as an AI chip rally lifted tech stocks, with Nvidia CEO Jensen Huang publicly dismissing “doomsday” fears about AI as unfounded.
LEARN HOW TO INVEST IN TECH STOCKS
Why People Were Already Nervous About an “AI Bubble”
Even before the sell-off, plenty of analysts were already uneasy.
Economists at Capital Economics published a widely discussed report on 10 September arguing that several market indicators were flashing warning signs last seen at the peak of the dot-com bubble in 2000- things like stretched valuations concentrated in a handful of tech giants, and forecasts that the biggest AI-focused “hyperscaler” companies could see their free cash flow turn negative in 2027 as they pour money into data centres and chips.
For the purpose of understanding the headlines: A “bubble” is when the price of an asset rises far beyond what its underlying business performance can justify, usually driven by hype and fear of missing out, and eventually corrects sharply. Whether AI stocks are in one is genuinely disputed- this isn’t a settled question, even among professional investors.
As if to prove the point about volatility, July had already seen Microsoft gain roughly $450 billion in market value in a single day, only for Apple to lose around $360 billion the next- the kind of swing that makes seasoned investors sit up and pay attention.
LEARN HOW TO INVEST IN TECH STOCKS
The Amodei Warning That Tipped Stocks Over
Against that already-jittery backdrop, Anthropic CEO Dario Amodei published a lengthy essay over the weekend of 12-13 September arguing that AI companies need to “pace” how fast they build more powerful systems, so that safety work can keep up. What made it a genuine market-moving event was that OpenAI’s Sam Altman and Tesla’s Elon Musk- companies and people who are normally fierce competitors- publicly agreed, with Altman warning about the risk of losing control over AI’s pace of development.
Investors read this as a signal that the breakneck AI investment boom might not continue at the pace priced into share prices. The result was a sharp sell-off on Monday 14 September: chip and infrastructure stocks fell around 6% in a single session, with some AI-linked names down 3-4%. Interestingly, some cybersecurity stocks actually rose, as investors rotated money within the tech sector rather than abandoning it altogether.
That same week, the US Federal Reserve added to the pressure, raising interest rates by a quarter point to a range of 3.75%-4% on 16 September- a hawkish move that tends to weigh more heavily on growth and tech stocks, which rely on future earnings that are worth less when borrowing costs rise.
The Bounce-Back: How Fast Markets Recovered
Selloffs driven by sentiment rather than a change in a company’s actual business often reverse quickly, and that’s roughly what happened here. By 16 September, chip stocks including AMD were already clawing back losses — AMD shares rose around 3.5% on the day, helped along by the company reiterating bullish guidance for its data-centre sales and maintaining a “Strong Buy” rating from the analysts who cover it.
The recovery gathered pace over the following days. By 21 September, the Nasdaq had climbed to a fresh record high, with chip names like Broadcom, AMD, and Nvidia leading the rally. In a CBS News interview around the same time, Nvidia CEO Jensen Huang directly addressed the AI safety debate, dismissing existential “doomsday” fears about AI as “completely false” while still acknowledging that safety research is a legitimate ongoing focus for the industry.
Where Things Stand Now
As of today, AI and chip stocks are sitting at or near record highs again — a genuinely whiplash-inducing round trip in the space of about three weeks. But the underlying debate that caused the wobble hasn’t actually been resolved. Valuations are still stretched by historical standards, the safety-versus-speed argument within the AI industry is still very much alive, and interest rates are now higher than they were at the start of the month. A record high today doesn’t rule out more volatility tomorrow.
Why This Matters for UK Beginner Investors
You don’t need to have deliberately bought “AI stocks” to have felt this. If you hold a global tracker fund, an S&P 500 or Nasdaq-focused ETF, a growth-focused investment trust, or even a fairly standard workplace pension default fund, there’s a good chance a meaningful chunk of it is invested in the handful of companies at the centre of this story.
What to Do Next
- Check what you actually hold. Log into your platform or pension and see how much of your portfolio is genuinely concentrated in AI and big tech, you might be more (or less) exposed than the headlines suggest.
- Don’t make decisions based on a single day’s headline. Both the crash and the record-high rebound happened within days of each other — reacting to either extreme in the moment is a good way to buy high and sell low.
- Revisit your diversification. If this month’s swings made you uncomfortable, that’s a useful signal to check whether your portfolio is spread across enough regions, sectors, and company sizes for your comfort level.
- Keep investing regularly if that’s your plan. If you invest a fixed amount each month, volatility like this means your contributions buy more when prices dip- a genuine (if uncomfortable) advantage of sticking to a regular investing habit.
- Expect more headlines like this. The AI bubble debate isn’t going away, so brace for further swings rather than treating each one as a unique crisis- start small, be patient, and let your plan do the work.
The Bottom Line
September 2026 has been a genuine rollercoaster for AI stocks: bubble warnings, a public industry disagreement about safety versus speed, a sharp sell-off, a Fed rate hike, and then a rebound to fresh record highs- all inside about three weeks. If you’re a UK beginner investor, the headline lesson isn’t which side of the “bubble or not” debate is right. It’s that this kind of volatility is a feature of investing in fast-growing, high-expectation sectors, and the best defence is a diversified portfolio and a plan you can stick to through both the crashes and the record highs.
This article is for general information and education only. It isn’t financial advice, and nothing here should be taken as a personal recommendation to buy or sell any investment. The value of investments can go down as well as up, and you could get back less than you put in. If you’re unsure, speak to a regulated financial adviser.
Leave a Reply