Americans’ AI Anxiety: The Burden of Early Adoption

Americans' AI Anxiety: The Burden of Early Adoption

I was standing in a Singapore subway car when a stranger tapped through a prompt on their phone and smiled. A week later, I was in a Manhattan newsroom where every conversation landed on alarms and layoffs. You can feel two very different futures breathing in the same world.

I write to you as someone who watches patterns, reads surveys, and talks to people on both sides of that aisle. The Gallup–Microsoft study released this week — the first slice of a 140-country effort, covering 37 nations so far — helps explain why optimism and fear live so uneasily together.

Awareness runs high in places where AI is already part of daily life.

In Singapore, nearly everyone I saw had heard of AI and many were already using it on the go.

The Gallup–Microsoft data shows a global median of 81% awareness among adults and 43% who have actually used AI. Singapore sits at the top: 96% aware, 79% users, and 46% using it daily. At the other extreme is Malawi, where awareness is just 12% and daily use barely registers.

Is AI adoption higher outside the U.S.?

Yes and no. Adoption varies more by infrastructure, education, and access than by geography alone. China, Vietnam, and Singapore show high engagement and optimism; many lower-adoption countries simply lack the internet or devices to experiment. The map is patchwork, not a straight line.

The U.S. mood is one of worry, even as Americans interact with AI more than most.

At my desk, colleagues describe AI as a present friction rather than a distant possibility.

Gallup calls this the “Paradox of the Worried West.” In the U.S., only 36% of AI-aware adults think the technology will mostly help the country, and 74% say AI makes them feel worried — the highest share among surveyed nations. Close behind are the Netherlands (67%), Canada (64%), the U.K. (63%), New Zealand (61%), and Ireland (60%).

Why are Americans more worried about AI?

Part of it is exposure: wealthy economies employ many knowledge workers whose tasks are easier to automate. Gallup also found that U.S. workers who use AI daily or several times a week are more than twice as likely to think their job could vanish within five years. Add continuous headlines about job loss, runaway models, and data-center energy use, and the anxiety compounds. You feel both the promise and the threat — a double-edged sword.

Optimism often outpaces trust in the technology’s accuracy.

In Beijing and Lagos, people I spoke with more readily describe practical improvements than existential dread.

Across the 37 countries, majorities in 25 expect AI to improve their daily lives, and optimists outnumber pessimists in 29 countries when asked about national benefit. China leads with 93% of AI-aware adults saying AI will mostly help their country, followed by Vietnam at 91% and Singapore at 77%. Yet the median share who trust AI “completely or a lot” to provide accurate information is only 36%. Only Israel, Vietnam, China, and Nigeria report majorities who trust AI a great deal.

Can AI be trusted to be accurate?

People separate hope from evidence: they expect gains in health, education, or services while remaining skeptical of current outputs. That gap — optimism for impact but low trust in accuracy — is a signal: you want the benefits, but you want them on your terms.

Companies are trying to soothe fears while regulators and researchers call for restraint.

In feeds and advertising, you’ll see more AI cheerleading than uncomfortable conversations.

OpenAI, Anthropic, and others have publicly asked for slower development at the frontier even as they push products like ChatGPT into millions of hands. Business Insider reported OpenAI’s influencer strategy to make the chatbot feel useful and approachable. Politico’s Public First poll shows roughly half of adults in several Western democracies favor pausing further development to reduce risks. At the same time, institutions such as the IMF warn of broad labor exposure — for instance, an IMF estimate suggests around 77% of Singapore’s workers are highly exposed to automation risk — and that complicates how people balance hope and fear.

Companies trying to reassure users are like a lighthouse in a storm: visible, comforting, but not a full solution to the waves coming at the coast.

I’ve told you what the data says and what people say on streets from Singapore to New York. You’re left with a split scene: many outside wealthy Western countries are quietly optimistic; many inside those countries are loudly anxious. Which side will shape rules, products, and markets — and whose anxieties will be answered first?