Here’s what the World had to say about the AI economy
AI is reshaping economies faster than governments are preparing for it. To understand what people around the world actually want from that transition — not policymakers, executives or technical experts, but ordinary citizens — Windfall Trust and the Collective Intelligence Project conducted the Global Dialogues Survey.
More than 1,000 people across 60 countries were asked what they expect from AI's economic future, who they trust to govern its gains, and what they want those gains to do. The survey was designed recruited respondents across the economic spectrum — including those most exposed to AI's consequences and least likely to have been asked.
Key Findings
Jobs Are Being Lost to AI. New Research Across 60 Countries Finds Its Economic Gains Bypassing People Most Affected
NEW YORK, March 31, 2026 — A new survey published today of more than 1,000 people across 60 countries has asked ordinary citizens — rather than policymakers, executives or technical experts — what they want from AI's economic future and how they believe its gains should be governed.
Conducted by Windfall Trust and the Collective Intelligence Project, the Global Dialogues Survey is a human-led piece of research that arrives just as governments, multilateral institutions and the private sector have produced a surge of AI research focused on economic projections and corporate strategy — with the global views of workers and citizens largely absent from that landscape.
"People told us AI is improving their lives and disrupting their workplaces — often at the same time. What they share across 60 countries is an experience of a transition that is already underway, and a governance response that isn't," said Adrian Brown, CEO of Windfall Trust.
The Global Dialogues survey was designed and conducted by human researchers, and recruited respondents across the economic spectrum — including those most exposed to AI and the least likely to have adopted AI as a tool.
Where surveys of existing AI users tend to find optimism about AI’s potential, this one finds a public that has absorbed both the benefits and the costs — people who use AI and find it useful, and who also personally know someone who lost a job to it — and wants governance that keeps pace with both.
Ninety-eight percent of respondents encounter AI at least weekly, and 72 percent report a net positive effect on their lives. But experience of AI and confidence in its consequences are different things.
Sixty percent personally know someone who has lost a job to automation. Sixty percent believe AI will have a net negative impact on the availability of good jobs — not just their own, but the broader economy's ability to generate meaningful work. And only nine percent trust national governments to manage the gains AI is expected to generate.
Forty percent expect their own jobs to be automated within a decade. When offered a choice between a universal basic income and job guarantee programs if AI displaces work at scale, 52 percent chose jobs. This finding challenges the narrative, influential in Silicon Valley and some policy circles, that a world without work is something most people would welcome. The preference for good jobs over income support was consistent across regions and income levels, including among respondents already in financially precarious situations.
As part of the survey, respondents were asked to take part in a thought experiment where they were asked to manage a pool of money as a global institution and make decisions on its behalf.
The governance findings are among the most distinctive in the survey — and go well beyond what has previously been asked.
“We have seen people - normally those who have already enthusiastically adopted AI - be asked what they hope AI will do for them,” said Anna Yelizarova, Chief Operating Officer at Windfall, who co-designed the survey. “What tends not to be asked is who they trust to make sure the predicted benefits of AI actually reach them.”
Across all countries, respondents independently arrived at the same two-stage model: a small expert team to design any wealth distribution system, followed by representative community councils to control it once operational. Fifty-two percent favoured the first stage; 69 percent the second. Only 9 percent want distribution routed through national governments, with nearly half of all concerns centering on fears of corruption or money ending in the wrong hands. These findings reflect a broader public that includes those most exposed to AI’s economic consequences and least represented in the debate about its governance, not simply experts or existing decision-makers. Across the regions surveyed, respondents said the top three things they wanted from AI for the communities were: the creation and protection of jobs (21%); more accessible healthcare (21%); and better education (21%).
AI is projected to add between $13 trillion and $15.7 trillion to the global economy by 2030, according to major financial and economic institutions. The survey finds little public confidence these gains will be shared — only 40 percent of financially struggling respondents believe AI's benefits will reach them, against 59 percent of those who describe themselves as financially comfortable.
On distribution, respondents had clear views. Forty-six percent say AI's gains should be shared because AI is built on shared human knowledge — a property claim rather than a welfare argument. Sixty-seven percent want resources delivered directly to individuals. And across every region, respondents said reaching the most vulnerable should come first, even if it slows distribution down.
Men and women hold similar views on what they want from AI, but diverge on whether they expect to get it. Fifty-four percent of women say the risks of AI outperforming humans on valuable work outweigh the benefits, compared to 42 percent of men. Multiple independent studies suggest that wariness is well-founded — women's jobs face significantly higher exposure to AI-driven displacement than men's. That gap widens when questions turn to the future: women are less confident than men that good jobs will remain available, less confident that AI's gains will reach them, and more likely to say the risks outweigh the benefits when asked about AI's long-term trajectory.
Regionally there were also differences. Respondents in Sub-Saharan Africa and Southeast Asia showed the highest demand for AI-funded public services and among the lowest demand for direct cash deposits. They were also the most precise about delivery mechanisms, naming tools that already work at scale in their communities rather than proposing systems yet to be built.
“This is an extraordinary data set that reflects what the global public wants, not only from the present, but from their economic futures in an AI-diffused world. Nothing like this exists, and we should listen to it,” said Zarinah Agnew, Research Director of the Collective Intelligence Project.