March 2026

Two Realities at Once

People find AI useful and threatening at the same time

For most of the world, the debate over whether AI will affect the economy is already settled, and the public is living with the consequences of both its promise and its disruption.

• Daily Fixture: 98% encounter AI systems at least weekly and over half use it daily; 72% say it has noticeably or profoundly improved their lives, with information access and learning cited as the primary change by 78%.
• Watching Displacement: Six in ten personally know someone who has lost a job to automation, and over a third know several; 40% expect their own job to be automated within a decade.
• The Trust Gap: Only 9% trust their government to manage the gains AI is expected to generate. People are not rejecting AI, they are signaling that the institutions meant to manage its consequences are not set up to do so.

Work Over Income

People want work, but not for the reasons policy debates assume

When work is no longer economically necessary, a majority still choose it, and the reasoning reveals that work is defended as a source of meaning, not merely a paycheck.

• The Preference: 52% chose a guaranteed-jobs society over a guaranteed-income society (39%); jobs led clearly in every region with a substantive sample, with North America the lone near-even split (47% jobs, 41% income).
• Meaning, Not Money: Those preferring jobs return to identity, purpose, and social connection. Work is the structure through which life has meaning, not primarily a source of income.
• Contested Reasoning: What divides people is not the choice but the justification. The claim that guaranteed income would "breed laziness" drew agreement from 10% of Buddhist participants but 75% of respondents in Indonesia.
• A Meaning Deficit: Only 27% say their current job is "good" by their own definition, yet 67% say it makes a meaningful contribution. People are defending what jobs are supposed to provide, not the jobs they have.

Public Services, Private Distrust

People want public services but don't trust public institutions to deliver them

The sharpest tension in the data is between what people want AI wealth to fund and who they trust to hand it out.

• The Paradox: 57% want AI wealth delivered as free public services (cash came last at 10%), yet when asked how a global fund should actually deliver, 67% chose cash sent straight to people and only 9% backed government grants or services.
• Rooted in Distrust: 56% actively distrust their government while only 27% trust it to any degree; corruption in fund distribution was the single most-cited concern (49%), followed by government interference (32%).
• Who They Do Trust: Public research institutions (70%) and small businesses (52%) are trusted far more than governments or large corporations; AI companies sit in the middle, and more people agree than disagree that AI would make better decisions than elected governments about its own impact (38% vs. 28%).

The Moral Logic of Redistribution

Not compensation for harm, but a return on common inheritance

Why people believe AI wealth should be shared matters as much as whether they do, and the dominant rationale is a claim about ownership, not welfare.

• Shared Knowledge Wins: 46% say AI's benefits should be shared because AI is built on shared human knowledge, versus 32% who frame it as job-displacement compensation and just 13% who cite the data-labor argument.
• A Property Claim: A compensation framing produces conditional safety nets indexed to harm; a shared-knowledge framing says collective achievements should yield collective returns. People see AI as a collective achievement, not unpaid data labor owed back wages.
• Regional Flips: The shared-knowledge framing led across most slices but reversed in Europe (job displacement 41% to 39%) and MENA (job displacement 50%), reflecting acute regional employment anxiety.

Governing What Doesn't Exist Yet

People prefer a global board of trustees: expert, diverse, and not elected

Asked how a hypothetical global AI wealth fund should be run, respondents converged on a phased, need-first model that sidesteps the governments they distrust.

• Scale Changes the Answer: 52% preferred a small expert team when a fund is starting; 69% preferred a large representative council at global scale, a 30-point swing emerging independently across 60+ countries.
• Chosen for Expertise: Respondents want council members selected for expertise (48%) rather than through elections (34%) or government appointment (11%). Large, diverse, and qualified, but not conventionally elected.
• Need Over Efficiency: Majorities would prioritize the poorest first even when AI isn't the cause of their poverty (64%), and launch where need is greatest even if distribution is slower (65%). A general anti-poverty tool funded by AI wealth, not limited to AI's casualties.

Circumstances Shape Attitudes

A person's material situation consistently shapes how they see AI's future

Economic position predicts optimism, but one regional pattern cuts sharply against the expected grain.

• Comfort Breeds Optimism: 59% of financially comfortable respondents believe AI's benefits will reach them personally, versus 40% of those who are struggling; wealthier regions prefer less work over more money, while South Asian and Sub-Saharan African respondents tilt toward higher earnings.
• The Leapfrog Exception: Sub-Saharan African respondents are the most optimistic that AI will reach their region (80%), naming specific applications they expect to transfer (healthcare, education, business tools) alongside specific barriers that won't (local language, cultural relevance).
• Build the Systems: These respondents were also the most enthusiastic about AI tools for communities (19%), suggesting a preference for AI that strengthens the systems people depend on over expanded government programs or cash transfers.

Conclusion

The findings from this Global Dialogue describe a public sketching the outlines of a social contract that hasn't been written yet. A majority want work for the meaning, identity, and community it provides, not just income. They want public services, but not through governments they don't trust. They want AI's gains shared, not as compensation for harm, but as a return on collective inheritance. These are not preferences that fit neatly into a tax code or a fund prospectus. As AI is projected to add trillions to the global economy, the question is not whether those gains will materialize, but who they will reach, and whether the institutions we have are the ones that can deliver them. Right now, the public's answer to that second question is clear, and it is no.

Methodology

This Global Dialogue, conducted in collaboration with Windfall Trust, surveyed 1,041 participants across 64 countries on the economic future they want from AI. It is a structured global dialogue rather than a nationally representative poll, one of the first to combine deliberative, interactive methods with global scale, allowing participants not only to contribute their own views but to engage with and vote on the ideas of others.

Participation is limited to those with reliable internet access, recruited through Prolific, and skews young (67% under 35), urban, and digitally engaged. The survey aims to surface regional patterns and reasoning to inform policy conversations rather than to achieve country-level statistical precision. Representativeness is tracked via the Global Representativeness Index and published for each round. Full results, methodology, and open-source data are available through Windfall Trust and the GD8 data repository.