Meta Platforms Inc. (NASDAQ:META) and OpenAI are rolling out AI agents, but Americans may adopt them more slowly than consumers in China for a counterintuitive reason: existing services in the West already work pretty well.

Howard Yu, LEGO Professor of Management and Innovation at IMD Business School, told Benzinga the West’s biggest obstacle may be its own convenience.

"Unless it’s perfect, I’m not going to try the alternative. Because life is good," Yu said, describing consumers in Europe and the U.S.

Why China May Move Faster

Yu said one of China’s advantages is the amount of context already sitting inside platforms such as WeChat.

During a visit to Tencent Holdings Ltd. (OTC:TCEHY), Yu saw a WeChat-based agent scan medicine packaging and produce personalized guidance within seconds using information already in the platform. He said that context can make an agent useful without requiring the most advanced model.

Yu said parts of Southeast Asia are on a similar path. In countries such as Indonesia and Vietnam, information has historically been more scattered and coordination more difficult, giving consumers more reason to try new digital services. "Anything better is infinitely better," he said.

Worldpay’s 2026 Agentic Commerce Report found a stark adoption gap: 86% of Chinese consumers were willing to shop with an AI agent, compared with just 45% in the U.S.

Muse Shows Interest, Not Yet Trust

Meta’s Muse suggests Americans are eager to try AI agents. The app reached roughly 2.5 million U.S. downloads through Sept. 19, with its first 12 days on iOS slightly outpacing ChatGPT’s comparable launch.

OpenAI followed with Dots, persistent agents that can carry out longer-running tasks using their own cloud computers and connections to more than 4,000 apps.

But downloading an agent is different from trusting it to make a purchase or booking. Asked whether Muse and Dots had changed his view, Yu said adoption would move at "very, very different" speeds across regions.

Yu cited travel-industry research showing consumers in Asia and the Middle East were much more willing than those in the U.S. and Europe to let an agent book a trip end to end.

Yu called that divergence "not good news" for the U.S., which he described as moving toward "more and more of a fortress economy" as other regions move faster.

US Labs Still Favored to Lead

Prediction market traders expect U.S. labs to dominate the frontier-model race. On Polymarket, Google and Anthropic are nearly tied to have the best AI model at the end of 2026, at about 43% and 42%, respectively. OpenAI follows at roughly 9%, while Meta sits near 3%. The market has attracted about $2.4 million in volume.

Chinese labs remain far behind in the market. Alibaba is around 1%, while DeepSeek, ByteDance and other Chinese names trade below 1%.

But the market measures model performance, not adoption. Yu’s argument is that China could move faster on AI agents even as U.S. companies continue to build the most capable models.

Image: Shutterstock

Read Also: Steve Eisman Warns S&P 500 Is One Giant AI Bet: ‘I Own NVDA and I Worry’