People in a Dozen Countries Trust China More Than the US or EU to Regulate AI. Here's Why That's Less Surprising Than It Sounds.

Pew Research Center published a finding this month that reads, at first glance, like a genuine reversal in global opinion: across 12 countries, more people trust China than the US or the EU to regulate AI effectively. A separate, earlier Pew survey found the opposite, with the EU well ahead and China a distant third. Both findings are accurate. The explanation isn't that opinion shifted. It's that the two surveys asked different people.

43%

Median trust in China to regulate AI, across the new 12-country survey

35% / 34%

Median trust in the US and the EU, respectively, in the same survey

13,696

Adults surveyed across the 12 countries, part of a broader 42,151-person, 37-country project

What the New Survey Found

Pew surveyed adults in Bangladesh, Chile, Colombia, Ghana, Malaysia, Pakistan, Peru, the Philippines, Singapore, Sri Lanka, Thailand, and the West Bank and East Jerusalem between February and May 2026, asking how much they trust the United States, China, and the European Union to regulate AI effectively. China led on a median basis, 43% to 35% for the US and 34% for the EU.

The pattern wasn't uniform. China was the single most-trusted regulator in five places specifically, Bangladesh, Malaysia, Pakistan, Sri Lanka, and the West Bank and East Jerusalem. Singaporeans trusted China and the EU equally, at the top. Chileans and Thais rated all three regulators at roughly similar levels. In Ghana, Peru, and Colombia, trust in the US and China came out about even.

Why This Looks Like a Reversal, and Mostly Isn't

A separate Pew survey of 25 countries, most of them wealthier nations in Europe, North America, and East Asia, found nearly the opposite ranking: a median of 53% trusted the EU to regulate AI, versus just 27% for China, with 60% actively distrusting China's ability to do so. Americans were among the least trusting of China specifically, at 13%, and Japanese respondents even lower, at 7%.

The Actual ExplanationPew's researchers point to a simpler mechanism than a genuine shift in global sentiment: trust in a country's ability to regulate AI tracks closely with how favorably people already view that country overall. The new 12-country survey drew heavily from middle-income nations, where China is generally viewed more positively than it is in the wealthier countries that dominated the earlier 25-country survey. Swap which countries get asked, and the ranking swaps with it. Nobody's mind changed. The sample did.

Why the Timing Still Matters

Even with that caveat, the release lands at a genuinely relevant moment. The survey was fielded between February and May, well before Dario Amodei's pacing essay, Sam Altman's and Elon Musk's public agreement with it, and the wider industry debate over AI safety that WorkplaceAI has covered extensively this month. That the data happens to surface now, in the middle of a live argument about which governments and companies the public should trust to manage AI risk, gives an otherwise methodological finding real relevance it wouldn't have had on its original release date.

It also complicates any simple story about global consensus on AI governance. There isn't one. Which regulator a population trusts most depends heavily on which population you ask, and any organization operating internationally should expect meaningfully different public reactions to the same AI policy depending on where its customers or employees are located.

What This Means for Global Operations

If your organization operates across multiple regions, don't assume a single AI governance framework will land the same way with users or regulators everywhere. Trust in who should oversee AI varies by country far more than most global AI strategy documents account for.
Treat any single-country or single-region AI opinion survey with caution before generalizing it worldwide. This pair of Pew studies is a clean demonstration of how much a survey's country selection can shape its headline finding.
Watch whether US and EU trust levels among middle-income countries continue tracking general favorability toward those governments, rather than assessments of their AI policy specifically. If that pattern holds, AI governance trust will keep moving with broader geopolitics rather than with anything AI regulators actually do.

Sources: AI Pulse · Big Picture · workplaceai.ai. The 12-country survey, "Do People Trust China, the U.S. or the EU to Regulate AI?", Pew Research Center, published September 17, 2026, as reported by the South China Morning Post and Fingerlakes1.com. The earlier 25-country survey, "Trust in the EU, U.S. and China to Regulate Use of AI," Pew Research Center. Additional context: IndexBox, citing Pew researcher Moira Fagan via Euronews Next. Every figure above is attributed to its original source; none is a WorkplaceAI study.