83% vs 39%: Why Is China So Optimistic About AI? Three Key Reasons

As the US and China prepare for a new round of talks on artificial intelligence and other issues, a striking data point has emerged: surveys show that 83% of respondents in China believe AI products bring more benefits than harms, compared to just 39% in the United States. Faced with the same rapidly evolving AI technology, the emotional starting points of people in the two countries couldn't be more different.

Why are Chinese people so optimistic about AI? China tech expert Kyle Chan offered three key explanations in a recent interview. These reasons go beyond the technology itself — they touch on people's experiences with technology over the past few decades, and their trust in governance systems.

Reason One: Fundamentally Different Memories of Technological Change

The most direct reason is that China has experienced a completely different version of technological transformation. For many Chinese people, technology has been a pathway to a better life — the state has clearly recognized technology as the key to modernization, and ordinary people have witnessed within their own lifetimes how technology has changed the way they live.

Over the past two decades, China's tech rise has been accompanied by rising household incomes. Technological progress and improved living standards have occurred almost in lockstep, and this positive association has been deeply embedded in the public's collective memory.

In the United States, by contrast, the public narrative around tech development over the same period has been dominated by privacy scandals, the spread of misinformation, and the growing market dominance of a handful of giant corporations. When technological progress becomes associated with negative events, the public naturally approaches a new wave of AI with far more skepticism.

In other words, whether a country's people are optimistic about AI depends largely on what technology has done for them in the past.

Reason Two: The Reassurance of Stricter Regulation

The second reason lies in governance. The Chinese government has taken a relatively stricter regulatory stance toward AI. In areas such as deepfakes, labeling of AI-generated content, and the psychological dependency that AI chatbots may create, China has already introduced fairly specific rules.

Kyle Chan notes that China is one of the few countries in the world that has actually written AI rules into law. For some people, clear regulation brings a sense of reassurance — it signals that the government will not allow AI to grow unchecked and out of control.

However, this strong degree of control has a darker side. The same regulatory mechanisms that can constrain technological risk can also be used to censor political speech. The boundary between trust and control is always a question that requires careful weighing. This reminds us that the source of optimism cannot simply be equated with AI itself being safer.

China's existing AI regulatory framework includes the Provisions on the Administration of Deep Synthesis Internet Information Services (2022) and the Interim Measures for the Administration of Generative Artificial Intelligence Services (2023), which require generative AI services to complete security assessments and filings, and mandate that AI-generated content carry watermarks or prominent labels. This framework is among the earliest and most comprehensive AI-specific legislation globally. It differs from the EU AI Act's risk-tiered approach — China focuses more on immediate control at the content level, while the EU emphasizes systematic pre-deployment risk assessment. Understanding this context helps clarify the specific mechanism by which "regulation brings reassurance": it stems more from the government's rapid response to visible content risks than from systematic guarantees of underlying algorithmic safety.

Reason Three: Fundamentally Different Visions for the AI Future

Perhaps the most important reason is that the US and China hold fundamentally different imaginations of what an "AI-driven future" looks like.

In the United States, the conversation is highly concentrated on AGI (Artificial General Intelligence) — the idea that humans may soon build AI models capable of performing nearly every human task. This narrative naturally carries a sense of tension: what happens when AI becomes more powerful than humans?

For China, the focus is far less on this single race toward AGI and much more on how to integrate AI into the broader economy. The AI vision asks: where can we put AI to use right now?

These two ways of framing the question represent completely different emotional starting points. The AGI vision focuses on distant, sweeping risks, while the pragmatic application vision focuses on near-term, visible value. This doesn't mean China ignores risk — but it does change people's initial reaction when they think about AI.

AGI (Artificial General Intelligence) refers to an AI system that can match or exceed human performance across nearly all cognitive tasks, in contrast to the "narrow AI" already widely deployed today (which excels only at specific tasks). Leading US labs like OpenAI and DeepMind explicitly list AGI as their ultimate goal, which has filled American tech culture with end-of-the-world narratives about AI "singularity," machine consciousness, and mass unemployment. By contrast, Chinese official policy documents (such as the New Generation Artificial Intelligence Development Plan) position AI as infrastructure for improving efficiency in manufacturing, healthcare, and urban management — rarely invoking the AGI framework. The divergence between these two visions reflects their respective logics of technological mobilization: one builds anxiety around a "general intelligence race," while the other builds anticipation around "industrial application deployment."

The Real Source of Optimism: Experience and Trust

It's worth emphasizing that China's optimism does not prove that AI is safer there. What it actually reveals is where trust comes from — it is a combination of two factors: what technology has done for you in the past, and whether you believe the system governing it can keep it under control.

Kyle Chan argues that the US and China should at least coordinate on their approaches to AI regulation. The US can and should take more seriously some of the AI risks that China has already been paying attention to, and think about how to design corresponding oversight. For example, when it comes to AI's relationship with children, establishing norms around age restrictions and interaction patterns.

Beyond the competitive narrative of "whoever wins AI wins the future," this comparison of optimism may offer a different perspective: the public acceptance of technology has never been purely a technical question. It is equally a social question about historical experience and institutional trust.