To put this in context, if Musk were a country, he would arguably have been well within an index of nation-state trillionaires on June 12. Before him on a nominal GDP basis (in USD trillions) are the United States ($32.38t), China ($20.85t), then a drop to Germany ($5.45t), Japan ($4.38t) and the United Kingdom ($4.24t). India, France, Italy, Russia and Brazil round out the top 10.
According to prediction markets such as Kalshi, traders are betting that Meta chief executive Mark Zuckerberg has a 32% chance of becoming the world’s second trillionaire. The next figures most likely to reach 13-digit net worths include Nvidia CEO Jensen Huang and Dell CEO Michael Dell, according to predictions.
Musk’s SpaceX operates Colossus and Colossus 2, a massive supercomputer campus in Memphis, Tennessee. Built to power the company’s AI assistant Grok, the supercomputers represent a major infrastructure build-out. SpaceX plans to produce advanced AI chips at scale in a joint project with Tesla using modular architecture.
Companies building foundational AI models – OpenAI, Anthropic, Microsoft, Alphabet, Nvidia, Meta and others – are constructing platforms that could become as indispensable to the global economy as electricity or the internet.
The implications extend well beyond technology. They reach into politics, taxation, employment, education and ultimately the future legitimacy of capitalism.
New Zealand has barely begun that conversation. Cabinet ministers talk about using AI to lift economic growth. Christopher Luxon talks about having an AI tsar at the top of government.
Businesses talk about reducing costs. Public sector leaders see opportunities to streamline bloated administrative systems by using AI to reduce “head count” – that absurd term beloved by boards and CEOs, which dehumanises the staff who work for them.
Former Microsoft chief financial officer Chris Liddell, now on the Anthropic board, spent years arguing that New Zealand could not continue accepting mediocre productivity. During his earlier career at Microsoft, and later in senior roles at the White House, where he rose to become deputy chief of staff (policy) to Donald Trump, he saw first-hand how technology transforms economies that embrace it rather than fear it.
It used to be said that New Zealand’s prosperity depended on producing more value from every hour worked. Without sustained productivity growth, wages stagnate, governments struggle to fund health and education, businesses lose competitiveness and talented people leave for larger economies. The Treasury, the Productivity Commission and successive Governments have all diagnosed the same structural weakness.
AI offers the prospect of finally shifting that needle. But productivity is not an end in itself. It is a means to creating wealth. The question policymakers have yet to answer is deceptively simple: Who owns that wealth?
The economics of AI are brutally efficient. Every boardroom is asking on behalf of shareholders, “How much faster can we complete routine work?” Companies that ignore AI will almost certainly be outperformed by companies that embrace it. The commercial logic is overwhelming.
Yet the social consequences are far less comfortable.
Anthropic chief executive Dario Amodei has warned that AI could eliminate roughly 50% of entry-level white-collar jobs within five years. These are precisely the jobs through which graduates learn their professions and develop experience. Law firms, accounting practices, consulting businesses, banks, insurers, media organisations and government departments all rely on these roles as the first rung of the career ladder.
Pointedly, Amodei forecasts this could lead to 20% unemployment.
Dario Amodei, the founder and chief executive of Anthropic. Photo / Getty Images
When the people building the technology begin warning governments that disruption may arrive faster than expected, policymakers should listen – but this issue is not registering in our own lamentably asinine election debate.
The Industrial Revolution delivered extraordinary wealth, but it also produced brutal working conditions, child labour and deep inequality before governments responded with labour laws, compulsory education, pensions and public health reforms.
The digital revolution generated immense fortunes but also democratised access to information and created entirely new industries.
Artificial intelligence is different. Unlike previous technologies, AI is not merely replacing physical labour. It is replicating cognitive work once thought uniquely human. It learns from decades of accumulated expertise contained in books, legal opinions, engineering designs, medical research, journalism (like mine), software code and scientific discoveries.
Even in my own news media industry, algorithms now make choices that were once made by highly skilled editors.
Human knowledge built through publicly funded schools and universities, taxpayer-supported research, generations of professional practice and billions of hours of human endeavour is the raw material.
Yet the overwhelming financial returns may accrue to a relatively small number of companies that own the algorithms, the chips and the computing infrastructure. If that occurs unchecked, AI could become the greatest engine of wealth concentration in modern history.
Markets can absorb remarkable levels of inequality. Democracies cannot.
That is why New Zealand should look to introduce an AI Productivity Dividend before events force the issue. This term matters.
Forget about the universal basic income popularised by fledgling political leaders who would heap more taxes on cash-strapped citizens to fund a benefit which even billionaires would get. That is so last century.
An AI Productivity Dividend would recognise that, when AI generates extraordinary productivity gains by replacing labour, part of that new wealth should be reinvested in the society whose knowledge made those gains possible.
There are multiple ways this could be designed:
A sovereign AI investment fund, underpinned by levies on AI majors;Tax incentives that encourage businesses to redeploy workers rather than simply eliminate positions;Large-scale investment in lifelong learning and advanced technical education;Transition support for workers displaced by automation – like the US used to have with trade displacement assistance.
If AI-driven returns prove as transformational as many investors now believe, there is the capacity to unlock enormous revenues. There is also the capacity for people not to have to work so hard – to be paid for less while machines pick up the slack; in other words, a well-funded leisure class.
The AI revolution could strengthen capitalism by broadening opportunity. Or it could weaken it by concentrating wealth on a scale that ultimately undermines public trust in the market economy.
The challenge is ensuring that New Zealanders remain stakeholders in the wealth AI creates, rather than becoming spectators while that wealth accumulates elsewhere.
That requires governments to grow a spine when it comes to dealing with the Googles and Metas and Microsofts of this world. Not to mention the rest of the foundational AI model platforms.
So far, the evidence is underwhelming.
Catch up on the debates that dominated the week by signing up to our Opinion newsletter – a weekly round-up of our best commentary.