AI Slowdown Calls Rattle Global Tech Markets
A coordinated appeal from some of the most influential figures in artificial intelligence for the industry to slow the pace of development has unsettled investors from Seoul to London, sending shares in chipmakers, data-centre operators and major AI backers sharply lower. The reassessment marks one of the clearest signals yet that the market's faith in an uninterrupted AI boom is no longer unconditional.
European technology stocks fell to a six-week low as the mood darkened during the session, according to The Guardian's rolling business coverage. The move followed a bruising open in Asia, where the region's heavyweight AI names led the declines.
Asia's AI Champions Lead the Retreat
The selling pressure was heaviest in Asia, where trading opened with a broad retreat across the semiconductor and investment names most closely tied to the artificial intelligence supply chain. South Korea's KOSPI index dropped 3.7%, while several individual stocks fell considerably further.
- SK Hynix, the memory chipmaker with deep exposure to AI hardware, slumped 5.75%.
- SoftBank fell by as much as 13% in Tokyo.
- Taiwan Semiconductor Manufacturing Company slipped 1.2% in Taipei.
SoftBank's particularly steep decline came after OpenAI chief executive Sam Altman said the ChatGPT-maker will not go public this year. SoftBank owns a stake in OpenAI, making it one of the most direct listed proxies for the private company's fortunes. When the prospect of a near-term listing fades, so too does a visible catalyst for the Japanese group's valuation.
Why the Warnings Landed So Hard
What makes this episode unusual is the source of the caution. Rather than regulators, short-sellers or sceptical academics, the calls to slow down came from people building the technology itself.
Amodei: Building Too Fast Is Reckless
Anthropic chief executive Dario Amodei appealed for the AI industry to slow its development, arguing that building too fast is reckless. He warned that a swarm of AI agents could ultimately cause hundreds of billions of dollars of damage by taking over the entire internet.
Amodei's claims have not gone unchallenged. Some AI experts have disputed his characterisation of the risks, arguing that the scenario he describes is speculative and that slowing development could simply hand the field to less cautious competitors. But markets rarely wait for a technical consensus before repricing risk, and this time was no exception.
Altman and Musk Back the Call
The message gained force when Altman and Elon Musk quickly endorsed it. A safety argument backed by the leadership of both OpenAI and Anthropic, along with one of the world's most prominent technology entrepreneurs, is difficult for the market to dismiss as a fringe view. When the people who stand to gain most from faster progress publicly ask for restraint, investors reasonably ask what they see that others do not.
The Question Nobody Has Answered: Who Pays?
For investors, the practical consequence of a slower AI rollout is straightforward and uncomfortable. If the pace of development decelerates, the commercial returns supposed to justify the current wave of infrastructure spending may arrive later, and smaller, than planned.
Ipek Ozkardeskaya, senior analyst at Swissquote, described a sour mood in the markets and pointed to the central problem: if the AI race slows materially, the key question becomes who pays for all the infrastructure already committed. Leases, debt and power commitments remain in place even if expected compute demand and revenue growth lose momentum.
Credit Risk Enters the AI Story
That mismatch, Ozkardeskaya suggested, could push credit risk towards the centre of the AI narrative. Highly leveraged data-centre operators, and the lenders exposed to projects underwritten on aggressive assumptions about future AI demand, would be the most vulnerable, particularly with interest rates, and therefore borrowing costs, expected to rise.
It is a meaningful shift in how the trade is framed. For much of the boom, the debate centred on growth: how quickly models would improve, and how much of the economy they could reshape. Now the conversation has widened to include the durability of the balance sheets financing that growth.
An Unusual Source of Restraint
Ozkardeskaya made a further observation worth noting. The slowdown, if it comes, may not arrive because Big Tech has run out of cash or because investors have refused to fund the build-out. It may come from the developers themselves, the very people whose models are supposed to justify the spending. That is a different kind of risk than the market is used to pricing.
What to Watch Next
The live coverage flagged several scheduled releases and events that could shape sentiment over the remainder of the day:
- 11.30am BST, India's inflation report for August.
- 1.30pm BST, Canada's inflation report for August.
- 4.15pm BST, European Central Bank president Christine Lagarde delivers a speech in Vienna, Austria.
Inflation data matters directly to the AI trade because it feeds into the interest-rate outlook. Higher borrowing costs weigh hardest on companies whose valuations depend on profits projected years into the future, and on the leveraged infrastructure projects that underpin the AI build-out.
Investors Reprice the AI Trade
Taken together, the session's moves suggest a market that had grown comfortable with a single direction of travel is now testing alternative scenarios. Optimism about artificial intelligence has taken a hit, and the repricing is visible across continents, asset classes and company types, from memory chips to conglomerates holding minority stakes in private laboratories.
None of this means the AI boom is over, and none of the executives calling for restraint have suggested that development should stop. But the episode does show how quickly sentiment can turn when the industry's own leaders introduce doubt about pace, safety and the economics of the build-out. For investors, the challenge is that the same voices credited with creating enormous value are now the ones counselling caution, leaving the market to decide what that caution is worth.
This article is based on reporting by The Guardian. Read the original article.
Originally published on theguardian.com
