Chapter 14

AI, THE CONCENTRATION OF WEALTH: A CHALLENGE TO GLOBAL EQUITY

Humanity faces a structural paradox of the 21st century: while artificial intelligence (AI) and technology amplify productivity and value creation, wealth is becoming increasingly concentrated in the hands of fewer and fewer actors. This phenomenon is not only economic, but also social and ethical, reflecting a profound imbalance between the advancement of knowledge and equity. This article addresses this tension and proposes a universal technical and methodological protocol to standardize the measurement of the Gini coefficient , with the aim of strengthening international comparability and guiding more effective redistributive policies.

1. The Paradox of the 21st Century: Knowledge and Inequality

Humanity today faces a paradox that defines the course of the 21st century: while artificial intelligence (AI) and technology multiply productivity and value creation, wealth tends to concentrate in the hands of a few. This phenomenon, far from being a simple economic outcome, reflects a profound imbalance between the advancement of knowledge and social equity.

For centuries, economic power has accumulated in the hands of those who control the means of production, land, natural resources, or financial capital. But in the digital age, this concentration takes a new form: the control of knowledge and data . Large technology corporations now possess more information than many nation-states, and with it, they determine the behaviors, preferences, and even political decisions of millions of people.

However, the concentration of wealth is not an inevitable fate. It reflects how societies distribute access to knowledge, education, and opportunities. When quality education and technology are reserved for a select few, a cycle is perpetuated that marginalizes everyone else. AI can break this cycle if it is used as a tool for democratizing knowledge , and not as an instrument of control.

2. The New Nature of Capital: Digital, Scientific and Human

Historical and structural factors—inheritance, inequality in land ownership, regressive tax policies, corruption, and lack of social mobility—continue to weigh heavily. But what distinguishes the present is that new capital is not only monetary , but also digital, scientific, and human. Whoever controls algorithms, research, or innovation concentrates both economic and symbolic power.

Consequently, education becomes the most powerful path to equity . An education system that fosters creativity, critical thinking, and technological proficiency can level the playing field. Investing in knowledge is investing in justice.

3. Policies and Ethics for an Equitable Society

The role of the state and civil society is fundamental. Progressive tax policies, regulation of technology monopolies, promotion of local entrepreneurship, and public transparency can reduce inequalities. But above all, a shared global ethic is needed , one that recognizes that the prosperity of a few cannot be built on the exclusion of many.

AI, when properly guided, can become a powerful ally for equality . Thanks to it, it's possible to design personalized education systems, optimize resource allocation, combat tax evasion, and ensure fairer and more transparent decision-making. The key lies in using technology consciously and responsibly.

History shows that civilizations thrive when they achieve a balance between material progress and collective well-being. Today, that balance depends on our ability to humanize artificial intelligence and put knowledge at the service of everyone.

In this new century, the real challenge is not accumulating more wealth, but distributing opportunities more equitably . Because a society is truly rich not when a few have a lot, but when everyone has the chance to grow, learn, and live with dignity.

4. Towards a Global Standard for Measuring Inequality

The development of effective redistributive policies requires comparable and methodologically consistent indicators across countries. In this context, the following approach is proposed as a universal standard for calculating the Gini coefficient :

4.1. Recommended Standard Concept

Use disposable income after taxes and transfers , expressed in purchasing power parity (PPP) and equivalent per capita income .

Reasons:

• It measures the real economic well-being of households (what they can actually spend or save).
• It allows for the evaluation of the redistributive effect of taxes and transfers.
• It facilitates the design of policies aimed at reducing effective inequality in living standards.

Furthermore, it is recommended to always publish the "market" Gini coefficient (pre-tax and pre-transfer). Comparing the two allows for quantifying the redistributive impact of public policies.

5. Recommended Formulas

5.1. Formula for Grouped Data (Lorenz Curve)

_{If the Lorenz curve with points (X k} , Y _k ) is available , where X _k is the cumulative proportion of population and Y _{k is} the cumulative proportion of income (with X ₀ = Y ₀ = 0 and X _m = Y _m = 1):

This is the trapezoidal formula over the Lorenz curve , standard for grouped data or percentiles.

5.2. Exact Discrete Formula (Individual Microdata)

When the incomes are ordered and ₁ ≤ y ₂ ≤ … ≤ y _n , where μ is the mean:

Or, equivalently, the general expression for absolute differences:

6. International Comparability and Equivalence

• Equivalence : Apply the square root rule of household size:andeq=andswhere s is the number of household members. It is simple, transparent, and widely accepted.
• PPP conversion : expressing income in purchasing power parity for international comparisons.
• Recommended unit of analysis : individuals equivalent per household.

7. Complementary Indicators and Decomposition

Measuring a single Gini coefficient is not enough. It is recommended to publish three Gini coefficients simultaneously :

1. Market Gini coefficient (pre-tax, pre-transfer)
2. Available or net Gini coefficient (post-tax, post-transfer)
3. Gini coefficient for consumption or wealth (if data is available)

Redistributive effect:

The Reynolds–Smolensky index or the Palma (9/40) index can also be used as robust complements.

8. Technical Practices for Universal Application

• Harmonize definitions of “disposable income”.
• Document the treatment of indirect taxes.
• Correct underreporting of high incomes through tax records.
• Publish confidence intervals and methodological sensitivity.
• Ensure transparency by publishing code and microdata.

9. Brief Protocol: Calculation of the Gini coefficient on Equivalent Disposable Income

Aim

To provide a clear, reproducible, and internationally comparable procedure.

Operational Steps

1. Collect microdata (income, taxes, transfers, household size, country).
2. Calculate disposable income:Rdisp=Rmarket−Direct taxes+Transfers
3. To equate:Rdispeq=Rdisps
4. Convert to PPP.
5. Clean and validate data.
6. Calculate the Gini coefficient using the exact discrete formula or the Lorenz curve.
7. Publish results with metadata and reproducibility.

10. Conclusion: Equity and Knowledge as Public Goods

In a world where AI is redefining work, value, and knowledge, measuring inequality is not just a statistical exercise, but an ethical and political decision. Adopting a universal standard for the Gini coefficient—based on equivalent disposable income, expressed in PPP, and with methodological transparency—will allow us not only to compare, but also to understand and transform the structures of inequality.

Just as AI can amplify inequality, it can also help us correct it. But this requires open data, inclusive education, and a global ethic of knowledge. Because in the new economy of the 21st century, equity is not an outcome: it's an architecture .