Summation - Chapter 4 Verifiable Thinking

By Scott Shields – Contributing Writer – Capitol Times Media - From Conversations and Material of Zhu Weisha. Learn more about Zhu Weisha here at Capitol Times Media's July Magazine Issue

The New Requirements for Economists in the AI Era

From the Revaluation of Productivity to the New Knowledge Structure of AI+Crypto

Introduction: Why New Facts Require New Theory

New facts do not automatically produce new theory. Theorists must enter the technological and institutional field to understand directly how AI changes cognitive productivity and how Crypto changes the foundation of credit.

Participant Economics does not abandon objectivity; it requires the researcher to enter the real process before forming judgments capable of explaining the new era.

Experts who do not understand this new knowledge structure of AI+Crypto will fall behind. AI changes productivity, while Crypto and verifiable finance change the credit foundation and production relations.

I. AI Is Not an Object Economists Can Comment on from the Outside

Is AI a bubble? This question is being debated across capital markets, the technology industry, and the economics profession. Rising valuations of AI companies, intense investment in data centers, and surging demand for chips naturally evoke memories of the internet bubble. This concern is not groundless.

But if AI is simply understood as a bubble, an even greater mistake will be made. A bubble is an asset-pricing issue; AI is a productivity issue. Asset prices can become bubbly, but a productivity revolution itself is not the same as a bubble. Railways, electricity, and the internet all went through bubbles, but after those bubbles burst, railways still transformed transportation, electricity still transformed industry, and the internet still transformed information dissemination and business organization.

The real challenge AI poses to economists is not whether they know the term "AI," but whether they understand that AI has already changed:

• Knowledge production

• Organizational efficiency

• Corporate value

• Labor structure

• Wealth distribution

• Credit systems

An economist who has not used AI deeply can easily stand outside AI and comment on it, while failing to see the real changes AI creates inside the workflow.

II. The Strengths of Traditional Economists Are Becoming Their Limitations

Traditional economists and financial scholars are skilled at studying money, interest rates, inflation, fiscal policy, corporations, capital markets, asset pricing, regulation, and risk. Their knowledge system was formed within the institutions of the industrial age and financial capitalism.

In that world:

• The corporation was the most important production organization

• Capital was the most important condition for expansion

• Labor was the most important production input

• Market prices were the most important signals

• Financial statements were the main window for judging corporate value

This knowledge remains important, but it is no longer complete. AI changes not one particular industry, but the production function of knowledge labor. Research, writing, translation, data organization, coding assistance, solution design, and decision analysis that once required many people and many days may now be completed by one person using AI in a very short period of time.

Much of AI's early value may not immediately enter the income statement, but it has already entered the workflow. It may not immediately appear as revenue growth, but it is already appearing as higher output per unit of time.

III. The Boundary of AI's Capabilities Can Be Seen Only by Deep Users

AI is first of all a production tool that knowledge workers can directly call upon. It is not a machine in the distance, but an external mind on the desktop. It is not merely an object of research, but a tool that participates in research itself.

Economists who do not personally enter this workflow will resemble people who study the internet without having used search engines, or people who study cryptocurrency without understanding Bitcoin nodes and private keys.

The value of AI does not exist abstractly in model parameters; it is reflected in concrete use:

• How to ask questions

• How to decompose tasks

• How to correct errors

• How to build workflows

• How to make AI understand one's own theoretical framework

• How to combine human judgment with machine capability

In intellectual originality, AI is still not as strong as human beings; in writing and organization, however, it is already stronger than human beings. Economists who do not understand this gap may overestimate AI's autonomous intellectual capacity, while also underestimating AI's transformation of the knowledge-production process.

IV. Asset Bubbles Must Be Distinguished from Productivity Revolutions

AI-related assets may contain bubbles, but AI itself is not equivalent to a bubble. This is the first distinction economists must establish.

Capital markets often capitalize on future returns in advance. Once the market believes that AI can change the economic structure, capital will flow early into chips, cloud computing, model companies, data centers, power infrastructure, and application platforms. Overvaluation, redundant construction, wasted investment, and narrative speculation will inevitably appear.

But an asset-price bubble does not mean that the technological revolution is false. After the internet bubble burst, the internet did not disappear. AI may follow a similar process: some companies may be overvalued, some capital expenditure may produce insufficient returns, and some business models may fail, but AI's transformation of knowledge labor, organizational efficiency, and machine decision-making will not disappear.

Economists must learn to separate the question of whether valuations are too high from the question of whether the technology is real.

V. Corporate Valuation Must Include AI Workflow Variables

In the AI era, corporate values cannot be judged only by traditional financial indicators. Financial statements remain important, but they are no longer sufficient.

New variables must be considered:

Traditional Variables New Variables

Revenue, profit, cash flow Has AI been embedded into workflow?

Balance sheets, market share Does management understand AI?

Brand, channels Can employees use AI as a production tool?

Return on capital, moats Has internal knowledge been structured for AI?

Management quality Have departments been restructured by AI?

These variables may not immediately appear in the income statement, but they will change the company's long-term competitiveness.

A small team that knows how to use AI may outperform a large team that does not. This will change the theory of the firm and corporate valuation.

VI. Individual + AI Is Changing the Production Unit

Traditional economics has long emphasized economies of scale at the enterprise level. AI changes this. AI does not simply replace individuals; it amplifies them.

For someone without intellectual accumulation, AI may be only a toy. But for someone with long-term accumulation, clear problem awareness, and judgment, AI becomes a productivity amplifier.

This means economists must rethink "individual productivity." In the future economy, not only companies, but individuals plus AI, small teams plus AI, and experts plus AI will become very important.

VII. AI Will Create New Questions of Wealth Distribution

Who owns the gains from AI? This is a question economists must answer in the AI era.

If AI becomes a general-purpose infrastructure, uses the long-accumulated knowledge of human society, and reshapes the labor market, society will naturally raise a question: should the excess returns generated by AI be enjoyed entirely by a small number of companies and shareholders?

This is not a simple anti-capital question, but a question about the distribution of general-purpose technology rents. When AI becomes as important as electricity, railways, the internet, or even monetary infrastructure, nations, societies, and the public will inevitably discuss:

• Public equity

• Sovereign wealth funds

• Government stakes

• Data rights

• Labor compensation

• Public dividends

One possible principle: AI platforms that have attributes of general-purpose infrastructure, use broad public knowledge, may generate extreme technology rents, and have systemic effects on the labor market should be studied for some form of public benefit-sharing mechanism.

VIII. AI+Crypto Is an Important Standard for Judging Whether Economists Understand the New Era

If AI changes productivity, then Crypto and verifiable finance change the credit foundation and production relations.

This is an important standard for judging whether economists still have reference value in the new era. An economist may not understand a specific model, chain, or company. But if he does not understand AI's change to productivity and also does not understand Crypto and verifiable finance's change to the way credit is produced, then his judgment will remain trapped in the old world.

AI Crypto/Verifiable Finance

Solves efficiency Solves credit

Makes knowledge production faster Provides new public verification structure

Expands individual capability Provides final landing point of chain proof

Lowers information production cost Returns key facts to verification structures

Economists who do not understand AI will underestimate the productivity revolution. Economists who do not understand Crypto and verifiable finance will underestimate the revolution in credit structure.

IX. AI Raises Productivity, But AI Does Not Provide a Credit Root

AI can improve efficiency, but AI cannot automatically provide credit. This is the key to understanding the relationship between AI and verifiable finance.

AI can write reports, but it can also generate fake reports. AI can assist auditing, but it can also assist fraud. AI can improve the efficiency of financial analysis, but it can also produce more complex deception.

Therefore, the stronger AI becomes, the less financial systems can rely only on "trust." When text, images, audio, video, reports, contracts, audit materials, and transaction explanations can all be rapidly generated by AI, society's demand for "verifiable facts" will become even stronger.

AI solves the problem of productivity; verifiable finance solves the problem of credit. AI changes knowledge production; verifiable finance changes the way credit is produced.

The future financial system will not become trustworthy through AI alone; it must anchor key facts to public credit roots through verifiable finance.

X. The New Knowledge Structure Economists Need in the AI Era

In the AI era, economists and financial scholars must add a new knowledge structure. Some capabilities are core and must be personally mastered, while others can be acquired through cross-disciplinary cooperation.

Core Capabilities:

1. Deep AI-use capability

2. Ability to reconstruct the production function

3. Ability to re-evaluate corporate value

4. Ability to distinguish asset bubbles from technological revolutions

Collaborative Capabilities: 5. Ability to design public-wealth and distribution institutions 6. Understanding of verification and credit roots

Among these, AI-use capability, production-function reconstruction, corporate-value re-evaluation, and the distinction between bubble and revolution should become basic capabilities for economists.

XI. Why Traditional Economists Are Likely to Fall Behind

Traditional economists are not falling behind because they are unintelligent, but because their knowledge framework was formed before AI and verifiable finance.

• They are used to studying the company, while AI is changing the boundary of the company

• They are used to studying labor, while AI is amplifying individuals and small teams

• They are used to studying capital, while AI is changing the relationship between capital and knowledge

• They are used to studying information asymmetry, while AI is changing information production itself

• They are used to studying regulation, while verifiable finance is transforming part of the object of regulation into verifiable facts

• They are used to studying subject-based credit, while public credit roots are providing a new credit foundation

Economists who fall behind are not those who fail to use new terminology, but those who continue to use old production functions, old theories of the firm, and old credit theories to explain a new world.

Conclusion: Economists Must Enter the New World, Not Observe It with Old Models

AI is not an ordinary technical tool, nor is it merely a capital-market narrative. It is changing:

• Knowledge production

• Corporate organization

• Individual capability

• Wealth distribution

• Credit risk

If economists only stand outside AI and comment on it, they will underestimate its change to productivity. If they look only at share prices and valuations, they will mistake a productivity revolution for a pure bubble. If they rely only on financial statements, they will miss the organizational changes occurring inside workflows.

The greatest requirement for economists in the AI era is to move from observer to participant.

They must personally use AI and understand AI. They must re-evaluate corporate value and understand the productivity of individuals and small teams. They must distinguish asset bubbles from technological revolutions. They must study the public distribution of AI's excess returns.

More importantly, they must understand that AI cannot automatically create credit, and that future finance will still need verifiable finance and public credit roots.

AI solves efficiency problems; verifiable finance solves credit problems. AI changes productivity; verifiable finance changes the production of credit. AI lets the world run faster; public credit roots let the world run more steadily.

The truly explanatory economics of the future will not be traditional economics with a few AI concepts added, nor the self-promotion of the technology industry. It must understand AI, Crypto, public credit roots, verifiable finance, and the institutional transition from probabilistic credit to verifiable credit.

In the AI era, the greatest risk for economists is not misjudging one company; it is using the knowledge structure of the old world to explain the new world.