I believe many friends have not yet read the article released yesterday by Citrini Research (THE 2028 GLOBAL INTELLIGENCE CRISIS). I made some modifications to the translation summary of AI to make it easier to understand.
(2028 Global Intelligence Crisis) Core logic reconstruction and problem setting: it is not a recession, but a structural replacement
The article constructs not a scenario of ordinary economic recession, but a systemic shock triggered by changes in the structure of production factors.
Core assumption:
• AI will achieve large-scale replacement of white-collar jobs between 2025 and 2027;
• The adoption of AI by enterprises is no longer just about efficiency optimization, but a survival strategy;
• Human labor loses its scarcity premium in the mid-to-high-end cognitive fields.
When 'intelligence' is no longer scarce, the pricing system of the economic system begins to misalign.
This is not a cyclical issue, but a change in the pricing foundation.
⸻
Two, the first layer of transmission: Employment → Consumption → Corporate Profits
The logical chain is very clear:
1. AI replaces white-collar jobs (programmers, financial analysts, legal, SaaS operations, etc.)
2. Income decline in high-income groups
3. Decline in consumption capacity
4. Corporate income is under pressure
5. Companies further expand AI use to cut costs
This creates a self-reinforcing cycle.
The key point is: The white-collar population contributes about 75% of consumer spending in the United States. Once their income structure is damaged, the impact is on the demand side, not the supply side.
Machines can produce, but machines do not consume.
The concept of 'Ghost GDP' points to a reality: Productivity rises, but cash flow cannot return to the household sector.
⸻
Three, the second layer of transmission: Balance sheet pressure
When the demand side contracts, financial structural issues begin to become apparent.
1. Real estate risk
• Approximately 13 trillion USD in housing loans in the U.S. are based on the assumption of stable income;
• Borrowers have high credit scores, but cash flow deteriorates;
• Default rates in high-income communities are rising.
The issue is not credit quality, but income stability.
2. Private credit and leverage structure
• The valuation of software companies is based on high growth and high gross margin assumptions;
• AI leads to SaaS homogenization and price wars;
• Declining income causes high leverage structures to lose support.
The risk of the private credit system lies in its non-transparency and structural nesting; once cash flow deteriorates, risks are difficult to price.
3. Insurance and reinsurance
Complex risk transfer structures are difficult to define in terms of responsibility when income collapses, increasing systemic uncertainty.
⸻
Four, the third layer of transmission: Fiscal and political tensions
Structural unemployment brings two direct consequences:
• Tax base contraction (decline in personal income tax)
• Increased spending (unemployment benefits increase)
Fiscal space is rapidly compressed.
Political divisions have intensified:
• Will there be a tax on AI income?
• Will wealth redistribution be implemented?
• Will a universal basic income be promoted?
The difficulty of the problem lies in: AI assets are highly concentrated, while labor income is widely dispersed. Changes in wealth structure will directly alter the political balance.
⸻
Five, the real core proposition
What the article truly discusses is not whether 'AI will lead to recession', but:
When human intelligence is no longer scarce, does the existing economic system still hold?
The modern capitalist system is based on three implicit premises:
1. Labor is the main source of income
2. Labor income drives consumption
3. Consumption drives corporate profits and asset valuations
If the first point is weakened, the latter two points will lose support.
This means that asset pricing models, tax systems, and credit structures all need to be re-evaluated.
⸻
Six, nature of risk: Structural rather than cyclical
Unlike 2008, this is not uncontrolled credit expansion;
Unlike 2020, this is not an exogenous shock.
This is a change in the production function itself.
The characteristics of structural problems are:
• The effects of traditional stimulus policies are limited;
• Monetary easing cannot solve income distribution issues;
• Policies lag behind technological advancements.
⸻
Seven, the practical significance of the article
The author does not assert that a crisis is inevitable, but proposes a risk framework:
If the speed of AI penetration is faster than the speed of institutional adaptation, economic and financial systems may become unbalanced.
Key variables include:
• Speed of AI replacement
• Labor retraining capability
• Wealth redistribution mechanism
• New demand creation capability
This is a timing issue, not a question of technological correctness.
⸻
Summary
(2028 Global Intelligence Crisis) is not about pessimism, but about structural deduction.
It raises a proposition worth serious consideration:
When production efficiency improves significantly, but the income distribution mechanism does not upgrade synchronously, the macro system may experience demand collapse and asset re-evaluation.
The real risk is not that AI is too strong, but that institutional adjustments are too slow.