Universal basic income, an idea traditionally championed by social philosophers and left-wing economists, has found unlikely backers in Silicon Valley as AI threatens to displace jobs. Eve Nicholson argues it will not work.

A decade ago, the largest poster ever created was placed on the Plainpalais plaza in Geneva. Printed across its surface was the provocation: "What would you do if your income were taken care of?"

That was the question posed to ordinary Swiss citizens in the summer of 2016, in a referendum proposing to transfer every adult in the country a minimum monthly income, regardless of how much they worked or whether they worked at all. More than three-quarters of voters rejected the proposal.

Yet a decade on, the seductive allure of universal basic income (UBI) has returned to the political limelight. Where earlier arguments for UBI were driven by a desire to tackle poverty or cut the cost of the welfare state, it is now heralded as the answer to an incoming onslaught: an AI-fuelled jobs apocalypse.

And rather than its usual brigade of social philosophers, left-wing economists and religious leaders, UBI is now backed by white-collar workers, venture capitalists, and even the wealthiest man on the planet, Elon Musk, who just a few months ago touted universal income as the "best way to deal with unemployment caused by AI." Sam Altman also put $14m into testing the idea a few years ago.

Advocates argue that as AI reduces the need for human labour and improves productivity, UBI is the best way of regulating and distributing wealth. Some also contend it is fair compensation for sharing our data in training "cognitive capitalism."

The idea has been tested in practice, not just theorised. Finland ran a two-year basic income experiment between 2017 and 2018, paying 2,000 unemployed people a monthly sum; recipients reported improved wellbeing, but employment effects were negligible. The results speak to the core dilemma UBI's advocates face: it can ease stress without rebuilding the place of work in people's lives.

Why UBI won't work in Britain

UBI has particular purchase in Britain, whose service-sector economy is highly exposed to AI relative to the US and other advanced economies, and Andy Burnham has previously described it as an idea "whose time has come."

But UBI is still not the answer.

Politically, it is impossible. Paying every UK adult the basic rate of Universal Credit – around £400 a month – would cost over £200bn a year, comparable to the entire NHS budget. Even if the whole benefit system were scrapped to finance UBI, broad-based taxes such as national insurance would have to rise, effectively giving with one hand and taking away with the other.

A popular alternative suggests funding UBI with a tax on AI gains. But there are political implications to relying on tech CEOs for a replacement welfare state, not least turning citizens into passive beneficiaries of a system in which they no longer participate.

Economically, AI-driven UBI could fuel inflation in sectors such as housing, health and care. While AI can drive "marginal costs toward zero" in software, digital content and administration, in housing, healthcare or energy infrastructure supply cannot expand at anywhere near the same pace, because of physical and political obstacles like land scarcity, planning regulations and professional licensing. And it is precisely into those sectors and the wider "relational economy" that people direct any extra cash. Raising people's nominal income would not produce abundance in these areas, but could drive prices up – exacerbating inequalities.

UBI is particularly seductive during periods of upheaval. Before Switzerland's 2016 vote, there was Paine's Age of Revolution "citizen's fund" and More's Renaissance "citizen's income." But rather than a quick-cash solution, such moments call for creative ways of navigating new relationships between labour, capital, citizen, state, work and meaning.

The alternative

There are signs this is happening. This April, Sam Altman said he "no longer believe[d] in UBI as much," and OpenAI released a white paper suggesting a swathe of alternative measures, including unemployment insurance, pathways "into human-centred work" and a Public Wealth Fund.

Altman, along with Microsoft and Anthropic, has funded a new initiative, RAISE US, centred on social growth in an AI economy. Its CEO, Gina Raimondo, likened UBI to "the end of America."

Sam Altman and others are right to change their tune. Rather than freeing labour from capital and distributing profit, AI-induced UBI would exacerbate inequalities, encourage greater unemployment, and be so overwhelmingly expensive that the only way to fund it would be to hand welfare provision over to unelected corporations.

Besides, people do not only work for money. Research from hundreds of cash transfer pilots over four decades finds that while UBI improves health and education, it does not boost labour market participation. In the face of AI, polls show most people would rather work. Beyond income, work provides purpose, relationships, organisation and mobility. UBI cannot replace that.

Eve Nicholson is a researcher and writer at The Week UK and editor of Running Dog Magazine.