Kamran Balayev is an international legal and policy expert, business leader, and former London mayoral candidate.
Britain often assumes that the biggest strategic questions will be settled elsewhere. Energy policy is shaped by global markets. Industrial strategy is constrained by fiscal reality. Artificial intelligence, surely, is a contest between Silicon Valley and Beijing.
That assumption is becoming harder to sustain.
AI is not merely another technology story. It is a question of economic power, military capability and geopolitical influence. The countries that shape its development will enjoy disproportionate advantages; those that do not will live with rules written by others.
Britain is not the obvious frontrunner. But nor is it a bystander.
Measured by raw spending, the hierarchy is clear. According to Stanford’s 2025 AI Index, American private AI investment reached $109bn in 2024, compared with Britain’s $4.5bn. Since 2013, the US has attracted roughly $471bn in private AI investment, against Britain’s $28bn. China, meanwhile, combines state backing, industrial policy and strategic patience.
Britain cannot outspend either.
But raw capital is not the only source of strategic influence.
Britain remains unusually well positioned in four respects: regulation, research, institutions and diplomacy.
The UK has a plausible opportunity to become, if not the capital of AI development, then at least a serious centre for AI governance. In strategic industries, the ability to shape rules can matter almost as much as the ability to deploy capital.
Its research base remains formidable. Oxford, Cambridge, Imperial and UCL continue to produce world-class work. Google DeepMind — one of the world’s most consequential AI laboratories — is headquartered in London.
More importantly, Britain has built institutions that many peers lack.
The AI Security Institute, established after the Bletchley Park summit, has given Britain something valuable: access. Its role in evaluating advanced AI systems alongside leading developers offers a degree of influence disproportionate to Britain’s economic weight. Regulatory credibility, in strategic industries, can matter as much as spending power.
That position should not be squandered.
The strategic stakes are becoming clearer by the month. AI systems are moving rapidly beyond chatbots into software engineering, scientific research, cyber capability and autonomous decision-making. Stanford’s latest index found that nearly 90 per cent of notable AI models in 2024 came from industry, up sharply from the previous year. The centre of gravity is shifting quickly.
This is increasingly a national security issue as much as an economic one.
Advanced AI will shape cyber operations, intelligence analysis, autonomous systems and military planning. The dividing line between commercial technology and strategic capability is disappearing.
That has implications far beyond productivity.
Eric Schmidt, the former Google chief executive, has warned that sufficiently advanced AI could create winner-takes-most dynamics, in which early leadership compounds through scale, compute and iterative improvement. Whether or not one accepts the more dramatic timelines, the broader strategic point is difficult to dismiss: in AI, leadership may be unusually difficult to dislodge.
As competition intensifies, so too will geopolitical risk.
It is no longer fanciful to imagine scenarios in which states facing technological disadvantage consider aggressive action to slow competitors. That is not a prediction. But the fact such scenarios are now discussed seriously tells its own story.
Britain cannot dominate this race. But it can still shape its terms.
First, it should treat standards-setting as a strategic objective, not a technocratic afterthought.
Much of Britain’s leverage lies not in building the largest models, but in helping determine the conditions under which powerful systems are tested, deployed and governed. Independent evaluation of frontier AI systems should become an international norm, not a voluntary courtesy.
Second, Britain should think more seriously about alliances.
No single country outside the US or China controls the AI supply chain. But allied nations collectively possess substantial leverage: Dutch semiconductor equipment, Taiwanese manufacturing, American hyperscale compute, British research capability and regulatory credibility.
Britain’s role is not technological autarky. It is coordination.
Its position across NATO, the G7 and Five Eyes gives it more influence than its balance sheet suggests.
Third, Britain must confront the infrastructure question honestly.
Serious AI capability requires compute, energy and talent. That does not mean attempting to replicate America’s hyperscale investment. It does mean ensuring Britain remains a credible place to build, research and deploy advanced systems.
Planning bottlenecks, weak grid infrastructure and sluggish strategic investment are no longer merely domestic irritants. They are competitiveness problems.
Talent matters just as much.
Britain continues to educate excellent researchers, only to lose many to better-funded ecosystems abroad. Some mobility is inevitable. Persistent talent leakage is not.
A country that trains top-tier AI researchers while failing to retain enough of them is effectively subsidising competitors.
All of this points to a broader political question.
British governments have often struggled to think strategically about emerging industries until advantage is already entrenched elsewhere. AI risks becoming another example.
But there is also a distinctly conservative case for acting.
The centre-right has traditionally understood something the current government too often overlooks: markets are powerful, but they do not automatically produce outcomes aligned with the national interest. Institutions matter. Security matters. Sovereignty matters.
Strategic technologies have never been governed by markets alone. America understands this. China certainly does. A serious conservative government should understand it too.
This is not an argument for state ownership, industrial nostalgia or futile attempts to outspend Washington. It is an argument for competent statecraft: setting rules, building credible institutions, investing where markets underprovide, and treating national capability as something worth preserving. That ought to be instinctive conservative terrain.
Britain retains genuine advantages. Unlike much of Europe, it has meaningful relationships with both American policymakers and frontier AI firms. Unlike smaller innovation hubs, it possesses diplomatic weight, strong legal institutions, defence credibility and research depth. It can often move more quickly than larger blocs when political will exists.
But advantages unused do not remain advantages for long. The question is not whether Britain can “win” the AI race outright. It cannot. The more relevant question is whether it intends to shape the environment in which that race unfolds.
If not, others will decide the standards, norms and dependencies that define one of the most important technologies of this century.
Britain has spent too much of the past decade reacting to strategic change rather than shaping it.
AI offers a chance to do otherwise. The machine is being built. Britain should help decide what it becomes.
Kamran Balayev is an international legal and policy expert, business leader, and former London mayoral candidate.
Britain often assumes that the biggest strategic questions will be settled elsewhere. Energy policy is shaped by global markets. Industrial strategy is constrained by fiscal reality. Artificial intelligence, surely, is a contest between Silicon Valley and Beijing.
That assumption is becoming harder to sustain.
AI is not merely another technology story. It is a question of economic power, military capability and geopolitical influence. The countries that shape its development will enjoy disproportionate advantages; those that do not will live with rules written by others.
Britain is not the obvious frontrunner. But nor is it a bystander.
Measured by raw spending, the hierarchy is clear. According to Stanford’s 2025 AI Index, American private AI investment reached $109bn in 2024, compared with Britain’s $4.5bn. Since 2013, the US has attracted roughly $471bn in private AI investment, against Britain’s $28bn. China, meanwhile, combines state backing, industrial policy and strategic patience.
Britain cannot outspend either.
But raw capital is not the only source of strategic influence.
Britain remains unusually well positioned in four respects: regulation, research, institutions and diplomacy.
The UK has a plausible opportunity to become, if not the capital of AI development, then at least a serious centre for AI governance. In strategic industries, the ability to shape rules can matter almost as much as the ability to deploy capital.
Its research base remains formidable. Oxford, Cambridge, Imperial and UCL continue to produce world-class work. Google DeepMind — one of the world’s most consequential AI laboratories — is headquartered in London.
More importantly, Britain has built institutions that many peers lack.
The AI Security Institute, established after the Bletchley Park summit, has given Britain something valuable: access. Its role in evaluating advanced AI systems alongside leading developers offers a degree of influence disproportionate to Britain’s economic weight. Regulatory credibility, in strategic industries, can matter as much as spending power.
That position should not be squandered.
The strategic stakes are becoming clearer by the month. AI systems are moving rapidly beyond chatbots into software engineering, scientific research, cyber capability and autonomous decision-making. Stanford’s latest index found that nearly 90 per cent of notable AI models in 2024 came from industry, up sharply from the previous year. The centre of gravity is shifting quickly.
This is increasingly a national security issue as much as an economic one.
Advanced AI will shape cyber operations, intelligence analysis, autonomous systems and military planning. The dividing line between commercial technology and strategic capability is disappearing.
That has implications far beyond productivity.
Eric Schmidt, the former Google chief executive, has warned that sufficiently advanced AI could create winner-takes-most dynamics, in which early leadership compounds through scale, compute and iterative improvement. Whether or not one accepts the more dramatic timelines, the broader strategic point is difficult to dismiss: in AI, leadership may be unusually difficult to dislodge.
As competition intensifies, so too will geopolitical risk.
It is no longer fanciful to imagine scenarios in which states facing technological disadvantage consider aggressive action to slow competitors. That is not a prediction. But the fact such scenarios are now discussed seriously tells its own story.
Britain cannot dominate this race. But it can still shape its terms.
First, it should treat standards-setting as a strategic objective, not a technocratic afterthought.
Much of Britain’s leverage lies not in building the largest models, but in helping determine the conditions under which powerful systems are tested, deployed and governed. Independent evaluation of frontier AI systems should become an international norm, not a voluntary courtesy.
Second, Britain should think more seriously about alliances.
No single country outside the US or China controls the AI supply chain. But allied nations collectively possess substantial leverage: Dutch semiconductor equipment, Taiwanese manufacturing, American hyperscale compute, British research capability and regulatory credibility.
Britain’s role is not technological autarky. It is coordination.
Its position across NATO, the G7 and Five Eyes gives it more influence than its balance sheet suggests.
Third, Britain must confront the infrastructure question honestly.
Serious AI capability requires compute, energy and talent. That does not mean attempting to replicate America’s hyperscale investment. It does mean ensuring Britain remains a credible place to build, research and deploy advanced systems.
Planning bottlenecks, weak grid infrastructure and sluggish strategic investment are no longer merely domestic irritants. They are competitiveness problems.
Talent matters just as much.
Britain continues to educate excellent researchers, only to lose many to better-funded ecosystems abroad. Some mobility is inevitable. Persistent talent leakage is not.
A country that trains top-tier AI researchers while failing to retain enough of them is effectively subsidising competitors.
All of this points to a broader political question.
British governments have often struggled to think strategically about emerging industries until advantage is already entrenched elsewhere. AI risks becoming another example.
But there is also a distinctly conservative case for acting.
The centre-right has traditionally understood something the current government too often overlooks: markets are powerful, but they do not automatically produce outcomes aligned with the national interest. Institutions matter. Security matters. Sovereignty matters.
Strategic technologies have never been governed by markets alone. America understands this. China certainly does. A serious conservative government should understand it too.
This is not an argument for state ownership, industrial nostalgia or futile attempts to outspend Washington. It is an argument for competent statecraft: setting rules, building credible institutions, investing where markets underprovide, and treating national capability as something worth preserving. That ought to be instinctive conservative terrain.
Britain retains genuine advantages. Unlike much of Europe, it has meaningful relationships with both American policymakers and frontier AI firms. Unlike smaller innovation hubs, it possesses diplomatic weight, strong legal institutions, defence credibility and research depth. It can often move more quickly than larger blocs when political will exists.
But advantages unused do not remain advantages for long. The question is not whether Britain can “win” the AI race outright. It cannot. The more relevant question is whether it intends to shape the environment in which that race unfolds.
If not, others will decide the standards, norms and dependencies that define one of the most important technologies of this century.
Britain has spent too much of the past decade reacting to strategic change rather than shaping it.
AI offers a chance to do otherwise. The machine is being built. Britain should help decide what it becomes.