The stakes are now in the trillions
Jensen Huang says artificial intelligence needs no new laws. He made the claim on 15 September. It is a simple statement. It is also a defence of a financial empire built on the promise of unstoppable, unregulated growth, a promise that has made his company, Nvidia, one of the most valuable corporations on the planet. This is not about philosophy. It is about finance.
The stakes are measured in the trillions. Nvidia’s stock market valuation has soared past one trillion pounds, a figure that places it in a tiny club of global titans alongside Apple and Microsoft. That number is not based on its current profits alone. It is a bet. Investors are gambling that Nvidia’s technology will form the backbone of a new industrial revolution, one powered entirely by artificial intelligence. They have bought into a story of exponential expansion, where Nvidia’s chips eventually sit inside every significant system, from financial markets and energy grids to medical diagnostics and military hardware.
This is why Huang fears new rules. Regulation introduces friction. It means delays for safety audits, immense compliance costs, and hard limits on what AI can be trained to do or be allowed to control. A regulator’s caution is an investor’s nightmare. Any serious attempt by the government to impose controls would puncture the growth story that justifies Nvidia’s colossal market price. The narrative would shift from boundless opportunity to managed risk.
For shareholders, that is a terrifying prospect. The trillion pound valuation is fragile. It rests on the belief that the AI boom will continue unchecked, accelerating every quarter. Huang’s intervention is an attempt to ensure that it does. He is not just a chief executive making a public statement, he is the guardian of a vast accumulation of capital trying to persuade governments in London, Washington and Brussels to stand aside and let the market run. His fortune depends on it. So does his company.
This is about more than money
The argument is not just about money. It is ideological. In Silicon Valley, speed is a religion, and its central commandment for decades has been to move fast and break things. This is the creed that built the modern internet. It prizes rapid iteration over cautious deliberation, launching products into the world to see what works and fixing the inevitable problems later. Jensen Huang is a high priest of this faith. His call for a light regulatory touch is the classic defence of a culture that believes progress cannot and should not wait for permission from governments. It is a very old argument. It is a very powerful one.
From this perspective, regulation is the great enemy of innovation. It is seen as a dead hand on the tiller. It means bureaucracy. It means delay. Imagine a small team of British engineers with a revolutionary AI concept, forced to navigate a labyrinth of compliance checks and pre approval panels before they can write a single line of production code. The venture capital dries up. The moment passes. The idea dies in a folder on a civil servant’s desk. This is the fear that haunts the tech industry, the belief that the administrative state is fundamentally incapable of matching the pace of technological change and will inevitably stifle the very creativity that fuels economic growth. They believe regulators kill momentum.
This is now a global race. The commercial argument is simple. If the UK, the European Union and the United States bind their own companies in red tape, others will not. China is building. Its top technologists in Beijing and Shenzhen are not waiting for ethical review boards to approve their next large language model, nor are they engaged in the same public soul searching over safety. They are sprinting. The nightmare scenario for Western policymakers is a future where the world’s most powerful artificial intelligence is developed and controlled by a geopolitical rival, creating an economic and strategic dependency that would reshape the global order. Losing this race is unthinkable. That is Huang’s leverage.
The argument forces a brutal choice upon governments. Move fast and accept the risk. Or move slowly and accept obsolescence. While Huang presents his case for freedom to innovate, a growing chorus of dissent is emerging from within the tech world itself. The BBC reports that his position is at odds with several current and former employees of major AI firms, people who have seen the technology up close. They are the insiders. They are worried. These experts now argue that the 'break things' mantra is dangerously irresponsible when applied to systems that could soon manage national power grids or global financial markets. The old Silicon Valley consensus is fracturing. The stakes are too high.
One mistake could break the market
One mistake could break the market. Imagine the City of London at nine o’clock on a Tuesday morning, not with traders shouting across a floor, but with silent servers processing transactions at the speed of light. Trillions of pounds are already moved by algorithms executing trades based on models of bewildering complexity, systems that are increasingly left to learn and adapt on their own with minimal human oversight. These are the systems that underpin global finance. They manage our pensions. Now imagine one goes wrong.
The failure would not be slow. It would be instant. An AI optimising a trading strategy could misinterpret a signal, or discover a novel but catastrophic flaw in market structure, triggering a cascade of sell orders that could evaporate billions in market value in seconds. This would be a flash crash without a human to blame, a digital pandemic spreading through the world’s connected financial exchanges before anyone in London, New York or Tokyo could even understand what was happening. Who is responsible? The law is silent. In the ensuing chaos, courts would be paralysed for years trying to determine if blame lay with the AI’s programmers, the bank that deployed it, or the firm that supplied the training data, leaving taxpayers to fund a bailout that could dwarf the 2008 financial rescue.
The risk is not confined to spreadsheets and stock tickers. Think about the power grid. An AI tasked with optimising national energy distribution for cost and efficiency might make a series of individually logical decisions that, when combined, create a fatal instability in the network during a winter peak. The lights would go out. The failure is catastrophic. A nationwide blackout would halt factory production, break supply chains, and leave critical infrastructure like hospitals and water treatment plants operating on emergency power. The economic disruption would be immediate and the insurance claims would be large enough to bankrupt entire consortiums, creating another shockwave that would ripple right back through the financial system.
This is the central, unanswered question of liability. The market rewards the promise of growth, but it has not yet priced in the cost of a single, large scale AI failure. It is a new kind of systemic risk. While companies like Nvidia count their revenue in the hundreds of billions, the potential damage from the misuse or malfunction of their technology is calculated in the trillions. The argument is not about stopping progress. It is about who pays when the bet goes wrong.
Nvidia is not a normal company
Nvidia is not a normal company. Its scale is vast. The company's stock market valuation recently passed £2.5 trillion. That figure makes it more valuable than technology giants like Amazon and Alphabet. It is larger than the entire combined value of every company on the London Stock Exchange. This is a new kind of corporate power. It is built on silicon chips. The sheer financial weight of the company gives Jensen Huang a voice that prime ministers must listen to, whether they are in London or Tokyo. He runs a new superpower.
The company’s dominance comes from one place. It makes the essential hardware for artificial intelligence. These are graphics processing units, or GPUs. They are the engine. Originally designed to render realistic visuals for computer games, these chips proved uniquely capable of performing the millions of parallel calculations needed to train large AI models. Nvidia had a huge head start. It now controls an estimated 80 per cent of the entire market for the data centre GPUs that power services like ChatGPT. This is not a competitive market. It is a near monopoly on the single most important component of the twenty first century’s defining technology, a position of leverage that John D. Rockefeller would have recognised.
This market control generates astonishing returns. In the first quarter of 2024, Nvidia reported revenues of £20.5 billion. That represents a 262 per cent increase from the same period a year earlier, a growth rate that is simply unheard of for a company of its size. The profits are enormous. The reason is the price of its products. A single one of its flagship H100 chips, the workhorse of the current AI boom, can sell for more than £25,000. A major data centre built by Meta or Microsoft might require tens of thousands of these units, creating orders worth billions of pounds for a single customer. Every company building AI needs these chips. And Nvidia is the primary supplier.
This context explains Jensen Huang’s resistance to new rules. Any regulation is a threat. New laws could mandate safety testing, impose performance caps, or force interoperability standards that would allow customers to mix Nvidia hardware with chips from rivals like AMD or Intel. This would weaken its hold. Any of these developments would slow down the revenue machine that has added more than a trillion pounds to the company’s valuation in the last year alone. Nvidia's stock price is not based on its current profits, as huge as they are, but on the promise that this explosive growth will continue for years. Regulation is the single biggest risk to that narrative. It threatens the entire financial architecture of the world’s most valuable company, a business that has become so powerful it can tell governments to leave it alone.
Governments are caught in the middle
Regulators are trapped. They face a choice between two equally perilous outcomes, either stifling a technology that promises generational economic growth or allowing an unregulated force to embed itself so deeply within the national infrastructure that a single failure could trigger a systemic collapse. The potential rewards are national. The potential costs are civilisation wide. A flash crash triggered by autonomous trading algorithms, or a failure of the automated logistics networks that supply supermarkets, could cause economic damage far exceeding the 2008 financial crisis. The stakes are enormous. They know this.
In London, the government tries to walk a middle path. It hosted the world at Bletchley Park for a summit on AI safety. Yet it resists hard legislation. The political calculation is that the United Kingdom can become a global hub for AI development by offering a lighter regulatory touch than its rivals, attracting billions in investment from companies that might otherwise build their European headquarters in Dublin or Amsterdam. It is a huge gamble, betting the country’s economic future on the hope that innovation and safety can coexist without binding legal force.
Brussels has taken a different path. The European Union’s AI Act, a sprawling piece of legislation, categorises systems by risk and imposes severe obligations on 'high risk' applications in areas like finance and energy. This is a direct challenge to the Silicon Valley model. For companies, non compliance could mean fines of up to €35 million or 7 per cent of global turnover, penalties large enough to get the attention of any boardroom. This is a rulebook. It is precisely what Jensen Huang is arguing against, a framework that prioritises safety over speed and subjects corporate decisions to government oversight.
Washington remains divided. Competing visions are battling for supremacy within Congress and the White House, with some lawmakers pushing for urgent federal rules while others, wary of handing a technological advantage to China, argue that the industry should be allowed to set its own standards for now. This indecision creates its own risk. It allows powerful new systems to become indispensable before any safety net is built. Every official in these capitals understands the prize, an economic dominance unseen since America’s post war boom. They also understand the penalty for getting it wrong. The pressure is immense. The clock is ticking.
The battle lines are drawn
The next six months will prove decisive. All eyes are on the other giants. Will the chief executives at firms like Microsoft and Google publicly endorse Jensen Huang’s free market position, or will they break ranks? A rival calling for rules could see a strategic advantage, positioning their own company as the responsible partner for nervous governments. Their statements will matter. The markets will listen.
Watch Nvidia’s share price. It is the most important number. The stock’s phenomenal rise has been built on a promise of near limitless growth, a narrative that new regulation would instantly puncture. Its daily movements are now a high frequency poll on investor confidence, a verdict on whether Huang's vision of an unregulated future is credible or just a fantasy. A sustained slide in the company’s trillion pound valuation would show that capital is betting against him, long before a single new law is passed.
Ultimately, the power rests with politicians. The critical moment will be the publication of the first serious draft legislation in either the United Kingdom or the United States. That is the real test. Such a document would transform the entire debate, moving it from abstract summits at Bletchley Park to the cold, hard reality of compliance departments and legal budgets. It would define liability. It would set penalties. Huang has placed his bet on innovation without restraint. Governments must now decide if they will call it.
Sources. BBC News Business: Nvidia boss says AI 'doesn't need new laws' as safety concerns grow. Sky News Technology: No need for new laws on AI, says Nvidia boss.
Analysis. Drafted with AI assistance from the sources listed above and reviewed by an editor before publication. Jnews links to the organisations it writes about.
