A new report from China Asset Management Co. (ChinaAMC) has exposed a troubling disconnect in how the world's tech companies approach artificial intelligence. While 92% of China-listed tech companies mention AI in their sustainability reports, only 22% have actually established specialized AI governance frameworks to manage the technology responsibly. This gap between high AI adoption and low governance oversight is not unique to China; it reflects a global phenomenon that regulators, investors, and companies are only beginning to address.

What Is the AI Governance Gap, and Why Should You Care?

The ChinaAMC report analyzed the 2025 environmental, social, and governance (ESG) reports of China's STAR 50 index constituents, a group of 50 leading tech companies. The findings paint a picture of an industry racing ahead with AI deployment while leaving critical safeguards behind. Among these 50 firms, 49 mentioned data security, privacy protection, and cybersecurity, and 48 touched on "science ethics." Yet only 11 explicitly discussed "AI ethics," "responsible AI," or "AI risks". A parallel study by UNESCO found similar governance gaps globally, suggesting this is not a regional problem but a worldwide challenge.

Why does this matter? As AI systems become embedded in everything from customer service to financial decision-making, the absence of formal governance structures means companies may be deploying powerful technologies without clear accountability, risk assessment, or ethical guardrails. This creates exposure not only for the companies themselves but for their customers and the broader economy.

How Are Companies and Regulators Addressing AI Governance?

The report identified three distinct regulatory approaches emerging across the world's largest economies:

• China: Balances AI development with security through agile legislation and rapid iteration, allowing the country to adapt rules as the technology evolves.
• European Union: Has built a stringent regulatory framework centered on risk classification, requiring companies to assess and mitigate AI-related harms based on severity.
• United States: Is experiencing a tug-of-war between federal deregulation efforts and tightening state-level oversight, creating an uneven regulatory landscape.

Meanwhile, investors are beginning to push back. The number of AI-related shareholder proposals in the U.S. stock market rose from 16 in 2023 to 26 in 2025. These governance resolutions garnered an average of approximately 30% support from independent shareholders, compared with only 16% average support for general environmental and social proposals over the same period. This suggests that investors view AI governance as a material risk worthy of board-level attention.

What Role Does Green Computing Play in Responsible AI?

One of the report's most significant findings concerns energy efficiency and green computing. Leading enterprises have widely incorporated continuous optimization of computing energy efficiency into their client contracts as a hard requirement. Green computing is shifting from a "nice-to-have" benefit to a mandatory "barrier to entry" for companies competing in the AI space. Although still in its infancy, the direction of this trend is certain, offering long-term structural opportunities for the industry.

This shift reflects growing awareness that AI's environmental footprint is not a peripheral concern but a core business issue. As data centers consume more electricity to power AI training and inference, companies that can demonstrate energy-efficient operations gain competitive advantage. Energy efficiency is becoming inseparable from responsible AI governance.

"Ultimately, we believe that the true impact of AI on ESG depends entirely on how the technology is designed, deployed, and governed," said Shirley Xu, ESG research head of ChinaAMC.

Shirley Xu, ESG Research Head at ChinaAMC

How Can Companies Build Responsible AI Frameworks?

ChinaAMC is actively constructing a "Responsible AI" evaluation framework to systematically assess tech companies across three dimensions:

• Governance Structure: Does the company have clear policies, oversight mechanisms, and accountability for AI deployment across the organization?
• Risk Identification and Assessment: Can the company identify potential harms from its AI systems, including bias, privacy violations, and unintended consequences, and does it have processes to measure and mitigate these risks?
• Positive Externalities: Is the company using AI in ways that create broader social or environmental benefits, such as improving energy efficiency or enhancing accessibility?

The report also highlighted that the consequences of AI adoption are not determined solely by technical boundaries. Corporate strategic attitudes play an equally vital role. When enterprises position AI as a "multiplier of employee capabilities" rather than a replacement tool, they can drive workforce empowerment and job restructuring, achieving a long-term win-win for labor value. This suggests that responsible AI governance extends beyond risk management to include how companies integrate AI into their workforce strategies.

What Does This Mean for the Future of AI?

The ChinaAMC report was launched during the World Artificial Intelligence Conference (WAIC 2026), which focused on "Responsible AI" as a central theme. The timing is significant. As AI systems become more powerful and more widely deployed, the gap between adoption and governance cannot persist indefinitely. Regulators are moving to close it, investors are demanding accountability, and leading companies are beginning to recognize that responsible AI is not a compliance burden but a competitive necessity.

The report's findings suggest that the next phase of AI development will be defined not by raw capability but by trustworthiness. Companies that establish robust governance frameworks, prioritize energy efficiency, and transparently address AI risks will likely emerge as industry leaders. Those that continue to treat AI governance as an afterthought may face regulatory action, investor pressure, and reputational damage. The window for voluntary action is closing, and the companies that move first may set the standards for the entire industry.