The AI Backlash Won’t Stop the Build-Out, It Will Just Move It
Why a more hostile US political climate could end up spreading AI investment around the world, and why some would argue the volatility is a buying opportunity.
With the Democrats routing Trump and taking control of both houses, a familiar argument is gaining ground: that artificial intelligence is about to be moderated in the United States. Data centres, grid access, energy consumption and the frontier labs themselves are all expected to face tougher scrutiny, and the conventional conclusion is that this must be bad news for the AI trade.
There is a different way to read it. Regulation tends to redirect capital rather than destroy it. If one were to ask where AI gets built in a world where the US becomes a more difficult place to build it, the answer is likely to be: everywhere else that wants it.
The capital is waiting in the Gulf
Qatar and the UAE have what AI infrastructure needs most: vast pools of sovereign capital and abundant, cheap energy. Both have spent years trying to convert hydrocarbon wealth into technology leadership, and both have the means to underwrite data centres at a scale few other jurisdictions can match. A tougher US environment hands them a stronger pitch: capital, power, speed and a welcoming regulator.
The talent is in London and Paris
The UK and France have deep research traditions, world-class universities and a steady pipeline of the engineers and scientists that frontier labs compete for. Both governments are keen to be seen as pro-AI and have every incentive to smooth the path for labs that choose to locate there. Talent is the scarcest ingredient of all, and it is already in place.
The land and power are in the north
Countries such as Finland and Canada offer cool climates, abundant clean electricity and plenty of space. These are close to ideal conditions for data centres, where cooling and power costs dominate the economics. Neither would find it difficult to build far more capacity than it has today, particularly if the demand arrived with sovereign or private backing.
Demonisation cuts both ways
There is also a political dimension that rarely features in the market debate. Many US technology leaders have been heavily criticised for their perceived closeness to Trump and his circle. If that association leaves them vilified at home, the incentive to look overseas grows. A company that feels unwelcome in one market has every reason to take its capital, its servers and its ambition to markets that are eager to receive them.
In that scenario, the effect of a US backlash is not to slow AI down but to internationalise it.
A backlash, not a verdict
There is no doubt that the current mood is hostile. The backlash against AI is real, loud and politically potent. But backlashes tend to be most intense when the technology is new, the benefits are diffuse and the costs are visible. Whether this one lasts beyond six to twelve months is doubtful, because the strategic logic is hard to escape. As evidence builds that other countries are catching up, the political calculus in Washington shifts as well. No government wants to be the one that legislated itself out of the defining technology race of the century.
And that is considering only the friendly nations. China and others are under no obligation to slow down for anyone’s electoral cycle, and the prospect of a rival power pulling ahead is precisely what tends to dissolve domestic political resistance.
The case for buying the dip
Some would argue now is a buying opportunity. Markets tend to price political risk at its peak, when headlines are loudest and uncertainty is highest. If the backlash proves temporary and the build-out simply relocates, the companies and infrastructure at the centre of AI could look cheap in hindsight. The geographic spread of AI investment, rather than hurting the leaders, may widen the total market they serve.
History offers a template. Hetty Green, the American financier known as the Witch of Wall Street, built a fortune through discipline: she was famed for holding large cash reserves and deploying them when panic forced others to sell. Her approach was never to predict the timing of the bottom, but to be prepared when fear made good assets available at a discount. The same instinct, loading up the truck when sentiment is at its worst, is what some investors will be applying to AI during this volatile period.
What could prove this wrong
No thesis is complete without its risks. Regulation could prove more durable and more global than expected. Other countries may take longer to build out meaningful capacity, or may adopt equally restrictive regimes of their own. Energy constraints and chip supply could limit expansion wherever it occurs. And valuations, even after a pullback, may still be demanding. Anyone acting on a view like this should size positions accordingly and diversify rather than concentrate.
This article is for general information and reflects a theoretical viewpoint only. It is not financial advice. Capital is at risk and past performance is not a guide to future returns. Do your own research or seek independent advice before investing.