Author: Zhang Feng

This article will explore how the promotion of standardized infrastructure reshapes industrial development and fundamentally changes the valuation logic of AI companies through the discussion and comparison of the characteristics of AI standardization in both countries.

In recent years, the rapid development of artificial intelligence (AI) technology has accelerated its transition from cutting-edge research in laboratories to commercialization across various industries. However, behind the technological frenzy, the valuation logic of AI companies has long been controversial, with market assessments often mixed with limitless hopes for the future. As technology applications enter deeper waters, risks and uncertainties become increasingly prominent, and policymakers, regulatory agencies, and investors begin to seek more robust and sustainable development paths.

In this context, regulators and industry leaders from both China and the United States have turned their attention to AI standardization and risk management. It is clear that standardization is becoming the key driver for the AI industry to move from "telling stories" to "focusing on practical work."

1. The United States AI Dictionary and Standardization Features of Risk Control

The U.S. Department of the Treasury recently released two new resources to guide the application of artificial intelligence in the financial sector: a shared AI dictionary and a financial services AI risk management framework (FSAIRMF). This initiative supports the President's AI Action Plan, which calls for clear standards, shared understanding, and risk-based governance to ensure the safe and responsible deployment of artificial intelligence.

“Implementing the President’s AI Action Plan requires not only idealistic statements but also practical resources that institutions can utilize,” said Deputy Secretary of the Treasury Derek Sulcer. “By establishing a common AI language and a customized financial servicesAI risk management framework, these deliverables help protect consumers and support responsible innovation.”

The United States has demonstrated a distinct characteristic of "pragmatism" and "collaborative governance" in promoting AI standardization, especially in its application in critical areas such as finance. The core idea is to translate macro national strategies into actionable guidelines for micro entities by constructing a common language and operational framework, thereby encouraging innovation while ensuring safety and stability.

First, the release of the "shared AI dictionary" marks a significant step forward for the United States in addressing fundamental challenges in AI governance. For a long time, terminologies in the AI field have varied significantly due to differences in academic backgrounds, application scenarios, and stakeholders. What technical developers refer to as "model interpretability" often points to different issues than what compliance departments focus on regarding "algorithmic transparency," and what business units understand as "decision logic." This inconsistency in terminology directly leads to inefficiencies in communication across departments and institutions, posing immense challenges for regulation. The AI dictionary launched by the U.S. Department of the Treasury aims to break this "Babel Tower" dilemma. By establishing a set of officially recognized, unified definitions for key AI concepts, capabilities, and risk categories, it achieves "synchronization" among regulators, technical experts, legal advisors, and business leaders. This not only helps financial institutions form a consistent understanding of AI risks internally but also provides clear metrics for external regulation, supporting more consistent and predictable implementation. This approach to standardizing the "language" itself reflects the U.S. emphasis on the fundamentals of AI governance and serves as the foundation for constructing a complex risk prevention system.

Second, the "financial services AI risk management framework" is constructed as a "manual" built upon a unified language. This framework does not start from scratch but skillfully adjusts and refines the macro AI risk management framework published by the National Institute of Standards and Technology to fit the specific context of financial services. This "tailor-made" approach represents the flexibility and precision of U.S. regulation. The core features of FS AI RMF include its full lifecycle and scalability. It covers the entire AI lifecycle from design, development, validation to deployment, monitoring, and updating, guiding institutions on how to identify AI application scenarios, assess potential risks, and embed accountability, transparency, and operational resilience in every aspect of AI deployment. Importantly, the framework is designed to be scalable and flexible, capable of adapting to the practical needs of institutions of various sizes and complexities, from startups to large multinational financial institutions. For instance, small fintech companies can use simplified tools in the framework to conduct preliminary risk assessments, while systematically important banks might need to establish more complex governance structures. This "tailored" design significantly increases the framework's adoption likelihood across the industry.

Finally, the promotion of AI standardization in the United States exhibits a clear characteristic of "public-private cooperation and multi-party governance." Whether it's the dictionary or the risk management framework, the development process is not a top-down dictation from regulatory agencies, but rather a collaborative effort involving public-private partnerships such as the Financial and Banking Information Infrastructure Committee and the AI Executive Oversight Group under the Financial Services Sector Coordinating Council. Positive evaluations from industry organizations like the Cyber Risk Research Institute also indirectly validate the framework's recognition within the industry. This multi-party participation model ensures that the outcomes of standardization reflect both regulatory concerns for safety and stability while also accommodating the industry's considerations for innovation efficiency and cost. The ultimate goal is to "support faster and broader AI applications in the financial sector," empowering the industry through enhanced cybersecurity and operational resilience rather than simply setting barriers.

2. Characteristics of China's AI Terminology and Risk Management Framework

China has corresponding official terminology standards and national-level AI safety governance/risk management systems to the U.S. Treasury's AI dictionary and AI risk management framework, having established a multi-tiered, full-process governance framework. Its core features can be summarized as "promoting development through standards and ensuring safety through regulations," aiming to establish rule dominance in the fierce global AI competition and ensure the healthy and orderly development of its domestic industry.

Its main components include the national standard "Information Technology - Artificial Intelligence Terminology" (GB/T 41867-2022) as the national core framework and the "AI Safety Governance Framework" (Version 2.0, 2025-09), supported by GB/T 46347-2025 "AI Risk Management Capability Assessment," providing organization-level AI risk management capability grading, assessment processes, and compliance guidelines. Meanwhile, the "Interim Measures for Managing Generative AI Services" (2023) clearly stipulates mandatory requirements for security assessments, filing, content review, and data compliance regarding generative AI services. Additionally, there are some best practice specifications like AI application risk management details released for key industries such as finance, healthcare, and education.

Compared to the pragmatic, industry-segmented incremental approach of the United States, China's construction of AI terminology and risk management frameworks demonstrates a stronger top-level design, faster promotion speed, and closer coupling with national strategies.

First, in terms of terminology standardization, China adopts a "systematic, forward-looking" construction strategy. Led by the National Standardization Management Committee, China is accelerating the construction of an AI standard system that covers fundamental commonalities, supporting technologies, product services, industry applications, and safety management at multiple levels. For example, the published "AI Terminology" national standard aims to provide a basic "common language" for the entire AI field.

Unlike the U.S.'s "shared AI dictionary" that focuses on a specific area of financial services, China's terminology standardization work is more global, attempting to clarify basic concepts, technical classifications, and development stages in the AI field from the roots. The advantage of this approach lies in its ability to provide a unified "foundation" for subsequent industry-specific standard development, effectively preventing contradictions and conflicts between standards across different industries, reflecting China's institutional advantage of "concentrating efforts to accomplish major tasks." Additionally, the process of developing these terminology standards closely tracks international cutting-edge trends, striving to integrate China's practices and understanding of AI into the international standard system, enhancing China's voice in global AI governance.

Second, in the risk management framework, China presents a prominent feature of "ethics first, safety as the foundation." China's AI governance framework is deeply influenced by the legal system of cybersecurity, data security, and personal information protection. Regulatory agencies such as the National Cyber Administration, Ministry of Industry and Information Technology, and Ministry of Public Security have intensively issued a series of normative documents targeting specific technologies and applications such as algorithm recommendations, deep synthesis, and generative AI, forming a multi-layered regulatory matrix. For instance, regarding generative AI services, China has taken the lead in implementing an algorithm registration and safety assessment system, requiring service providers to be responsible for the legitimacy of training data, fairness of algorithms, and authenticity of generated content.

This regulatory approach, compared to the U.S. FS AI RMF model which emphasizes internal governance and self-assessment of risks, is more prescriptive and grounded in bottom-line thinking. It clearly delineates the "red lines" of AI development, especially in areas like data security, ideological security, and protection of citizens' rights and interests, reflecting very stringent regulatory requirements. China's risk management framework more often manifests as an "external compliance constraint," driving enterprises to establish internal risk control systems to meet regulatory demands.

Finally, China's promotion of AI standardization is highly coordinated with industrial development and national strategic goals. Standardization is viewed as a key infrastructure for enabling AI to empower the real economy and achieve high-quality development. For example, in the financial sector, the development plan published by the People's Bank of China clearly requires strengthening the supply of AI finance application standards, covering aspects such as intelligent risk control, intelligent marketing, and intelligent customer service. These standards focus not only on risk prevention but also strive to enhance the efficiency and inclusiveness of financial services.

The underlying logic is that by standardizing technical interfaces, data formats, and assessment methods, the collaboration costs across the entire industry chain can be reduced, promoting the widespread application of AI technology in the financial sector. At the same time, the implementation of standards also provides a "philosopher's stone" for leading technology companies, promoting them to translate mature technical solutions into industry norms, thereby consolidating their market positions. This "promoting industry through standards" approach makes China's AI standardization process not just a regulatory tool but also an important engine for driving industrial upgrading and cultivating new quality productivity.

3. A Comparison of AI Standardization Infrastructure between China and the United States

Despite both China and the United States having deeply recognized the importance of AI standardization and actively taking action, the fundamental differences in political systems, market environments, innovation cultures, and regulatory philosophies lead to significant differences in the construction paths, core characteristics, and implementation effects of their AI standardization infrastructures.

From the perspective of top-level design and bottom-up drive, China's AI standardization is a typical "government-led, top-down" model. The national level has clear strategic planning for AI development, and standardization work serves as a key support for realizing this planning, coordinated by the National Standardization Management Committee, with various ministries collaborating in their respective fields. The prioritization of standard setting is highly consistent with national industrial policies and technological breakthroughs, possessing strong guidance and enforcement. The advantage of this model lies in its high efficiency and strong execution capability, enabling the rapid establishment of a broad standard system.

In contrast, the U.S. AI standardization is characterized by "market-driven, bottom-up" processes. The government's role is more about being a "facilitator" and "promoter," guiding the industry to spontaneously form consensus through the issuance of guidelines, frameworks, and best practices. Its standardization process emphasizes multi-party participation, consensus, and fully respects the innovative vitality and professional judgment of market entities. The development process of FS AI RMF is a typical example, and its results lean more towards "recommended guidelines" rather than "mandatory regulations." The strength of this model lies in its flexibility and adaptability, making it less likely to stifle innovation, though it may fall slightly short on consistency and speed of implementation.

In the core focus of their standard systems, there are also subtle differences between China and the United States. The AI standard system in China, particularly in risk management, highly focuses on "safety and controllability" and "ethical compliance." This originates from China's high regard for cybersecurity, data sovereignty, and social stability. Thus, standards often impose strict requirements on data legitimacy, algorithm fairness, content authenticity, and system accountability, and are often closely linked with higher laws such as the Cybersecurity Law, Data Security Law, and Personal Information Protection Law. Regulatory agencies tend to conduct preemptive or process-oriented regulation of AI applications through clear rules and filing and assessment procedures.

Conversely, the U.S. AI risk management framework, while also considering safety and fairness comprehensively, leans more towards a "risk-based" model of institutional self-governance. Its starting point is to help institutions identify, assess, and manage their operational, reputational, and compliance risks to support the realization of their business goals. It emphasizes that institutions should establish dynamic, continuous risk management processes based on their risk preferences and application scenarios instead of mechanically adhering to a fixed set of rules. This difference reflects the fundamental variations in regulatory philosophy between the two countries: China tends to use uniform rules to regulate market behavior to prevent systemic risks, while the U.S. places greater trust in the self-management capabilities of market entities.

From the perspective of the interaction between standards and industry, the Chinese model focuses on "pulling" industrial development through standards. Leading AI companies, especially top technology firms, often deeply participate in the formulation of national and industry standards. This not only reflects their technical strength but also serves as an important means for them to build industrial ecosystems and establish competitive advantages. Standards become important catalysts for technology diffusion and large-scale application.

In the United States, standards are more of a "summary" and "elevation" of industry best practices. FS AI RMF largely incorporates the risk management experiences accumulated by financial institutions and technology companies in practice. This model ensures that standards remain synchronized with industry frontiers and avoids the problem of standards lagging behind technological advancement. However, this may also lead to fragmentation in the standard system, requiring government-level integration and coordination.

In terms of international influence and compatibility, both China and the United States are committed to promoting their national standards internationally. With its vast market and strong industrial strength, China is actively exporting its standard concepts and practices through international standardization platforms like ISO/IEC JTC 1/SC 42. Meanwhile, the United States, relying on its traditional dominant position in the global technology field, finds its NIST framework and other "soft laws" enjoy strong global resonance and de facto influence. In the future, global AI governance is likely to form a complex pattern of competition and limited cooperation between the two major standard systems of China and the United States.

4. The Impact of AI Infrastructure Promotion on Industrial Development and Valuation Logic

Whether it is China's top-down systematic construction or the United States' bottom-up industry consensus, an undeniable fact is that the continuous improvement of AI standardization as a fundamental infrastructure is profoundly reshaping the development trajectory of the AI industry and fundamentally overturning the previous irrational prosperity that relied on "telling stories" to support valuations.

First, standardization greatly reduces the transaction costs and entry barriers of the AI industry, promoting the "ubiquitous" application of technology throughout the economic system. Unified terminologies and interface standards allow AI components developed by different enterprises to be flexibly assembled and deployed. This "plug-and-play" standardization model significantly accelerates the transition of AI technology from laboratories to factory floors, fields, and bank counters. The focus of industrial development will shift from "how to create AI" to "how to effectively utilize AI."

This means that companies with only algorithmic technology but lacking deep understanding and application scenario capability of vertical industries will face a reevaluation of their value. In contrast, those providers of "AI + industry" solutions that deeply understand industry pain points and combine standardized AI technologies with specific business processes to create significant business value will gain market favor.

Second, the establishment of risk management frameworks provides the market with a common metric for assessing the "health" of AI companies. In the past, risk assessments of AI companies were often vague and subjective. Now, whether it is the U.S. FS AI RMF or China's regulatory requirements in finance and cybersecurity, they provide concrete dimensions for assessing an AI company's sustainable operation capabilities.

Investors are beginning to pay attention to: Does the company's AI model have bias risks? Is the source of its training data legal and compliant? Does the model's decision-making process possess interpretability? Has the company established a risk management process covering the entire AI lifecycle? These previously overlooked "soft skills" are now becoming key factors determining a company's success or failure. A company able to provide efficient AI services while ensuring data privacy, algorithm fairness, and system safety is undoubtedly more resilient and sustainable, deserving a valuation premium.

Third, standardization and compliance requirements are becoming crucial selection mechanisms for survival of the fittest in the AI industry. Meeting the increasingly complex compliance requirements requires significant investment of human and financial resources. For startups, this creates a considerable "compliance threshold." Objectively, this favors larger, more resource-rich, and better-managed leading companies.

At the same time, standardization provides a basis for customers to select AI products and services. An AI product that has passed relevant national standard certification or follows internationally recognized risk management frameworks is more likely to earn customer trust. This trust based on standards will become an important component of branding, further solidifying the market position of leading enterprises. This indicates that future competition in AI will no longer merely be about technology and algorithms but will encompass comprehensive competition in governance capability, compliance capability, and brand reputation.

Ultimately, all these elements will lead to a fundamental shift: the core of AI company valuation is transitioning from "possibility" to "certainty." In the early stages of AI development, the market was keen on chasing stories that painted visions of a "future world." This "storytelling" logic supported a significant amount of early investments and high valuations, but it also created substantial bubble risks.

The improvement of AI standardization infrastructure is, in fact, a process of squeezing this bubble. It requires companies to break down grand visions into measurable, manageable, and verifiable specific indicators. A company's value will no longer solely depend on its founders' visions or the number of its papers at top academic conferences but will increasingly rely on healthy revenue growth, successful customer case studies, barriers of core technologies, effectiveness of risk management, and records of compliant operations.

In summary, although China and the United States are exploring different paths in AI standardization, they collectively point toward a clear future: AI is evolving from a technological gold rush into a mature industry with clear rules, infrastructure, and risk management. The release of the AI dictionary has eliminated communication noise; the implementation of risk management frameworks has defined the boundaries of action; the improvement of standardization infrastructure has built a sustainable ecosystem. In this grand context, the valuation logic of AI companies will undoubtedly undergo profound changes. Enterprises that can traverse conceptual fog and build safe, trustworthy, efficient, and truly valuable AI applications on solid foundations of standardization will emerge as the winners of the new era, while the once-prevalent pure "storytelling" logic will ultimately be abandoned by the market.

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。