Note to Readers. This is a Google NotebookLM AI generated analysis and audio podcast of articles by Carlo Iacono, Alberto Romero and others speculating and predicting on what might happen to jobs and the human economy in the next year or so.
1. The Human Intelligence Displacement Spiral
In hindsight, the signal was clear, yet the market mistook a secular structural headwind for a manageable cyclical one. By October 2026, the S&P 500 had crested at 8,000 and the Nasdaq broke 30,000, fueled by the conviction that AI-driven margin expansion was a permanent state of grace. This euphoria ignored the “Human Intelligence Displacement Spiral,” a negative feedback loop that would eventually trigger a 38% drawdown and a 10.2% unemployment rate.
The mechanical sequence was ruthless: step-function jumps in agentic AI capability allowed firms to protect short-term earnings by replacing high-wage human labor with machine intelligence. This triggered a destructive cycle: white-collar layoffs reduced aggregate consumer spending, which created further margin pressure, forcing firms to redeploy savings into even more AI compute to maintain output at lower costs. This improved AI capability further, initiating the next round of displacement.
To illustrate the socio-economic downshifting inherent in this spiral, consider the “Salesforce Case Study,” which became the archetype for the white-collar unwind:
Case Study: The Senior Product Manager In 2025, a Senior Product Manager at Salesforce earned $180,000 per annum, including health insurance and a 401k. Following the third round of AI-driven “structural efficiency programs” in 2026, she faced six months of unemployment. Eventually, she was forced to “downshift” into the gig economy, driving for Uber. Her annual earnings plummeted to $45,000—a 75% reduction in income. This transition from a high-wage professional role to a low-wage service role was replicated by hundreds of thousands of workers across major tech hubs.
This “overqualified labor flood” into the service and gig sectors compressed wages for the entire bottom 80% of earners, as the supply of human labor vastly exceeded demand in the remaining human-centric roles.
--------------------------------------------------------------------------------
2. The Phenomenon of ‘Ghost GDP’ and Economic Decoupling
The 2028 crisis gave rise to the concept of ‘Ghost GDP’—economic output that appears in national accounts and suggests a booming economy while failing to circulate in the human-centric real economy. By early 2027, productivity was surging, but the gains were flowing exclusively to owners of compute. Labor’s share of GDP, which stood at 56% in 2024, plummeted to a historic low of 46% by 2028.
Traditional GDP Indicators
Human Economy Reality
Productivity Growth: Record-setting real output per hour, driven by 24/7 AI agents.
Velocity of Money: Flatlined as income failed to route through human households.
Corporate Profits: Margins expanded to record highs as labor costs were deleted.
Real Wage Growth: Collapsed into negative territory and remained there for 18 months.
Nominal GDP: Boosted by massive AI compute and $200B/quarter data center CapEx.
Discretionary Spending: Withered; as the Citrini memo noted, “machines spend zero on consumer goods.”
This decoupling was driven by “machine-to-machine” commerce. In an economy where AI agents handle transactions, the circular flow of income is broken. Output no longer translates into taxable household income that can be spent back into the system by humans, leading to a withered consumer economy—which historically accounted for 70% of U.S. GDP.
--------------------------------------------------------------------------------
3. The Collapse of Intermediation and the Zero-Friction Economy
The crisis saw the disintegration of business models based on “habitual intermediation”—the monetization of human inertia and friction. AI agents, running 24/7, removed the consumer constraints of time and patience that trillions in enterprise value once relied upon.
The End of Habitual Loyalty: The “DoorDash moat” was built on the reality of humans being hungry, lazy, and clicking the first app on their home screen. AI agents, however, do not have home screens. They route orders to whichever “vibe-coded” alternative or restaurant-direct site offers the lowest fee and fastest delivery, compressing margins to zero.
Agent on Agent Violence: In real estate, AI agents equipped with decades of transaction data replicated the knowledge of human brokers instantly. Information asymmetry vanished, and buy-side commissions in major metros compressed from 3% to under 1% as transactions closed without human intervention.
The “Toll Booth” Collapse: Machines transacting with machines sought to eliminate the 2-3% card interchange rate.
Credit Card Interchange: AI agents began routing around Mastercard and Visa.
Stablecoin Settlement: Machine commerce shifted to near-instant settlement on Solana and Ethereum L2s, reducing costs to fractions of a penny.
Mono-line Issuers: American Express and Capital One saw their reward-based models gutted as the merchant subsidy from interchange fees evaporated.
--------------------------------------------------------------------------------
4. Financial Contagion: Private Credit and the Global Unwind
The crisis evolved from a tech shock into a systemic contagion through a “daisy chain of correlated bets” on white-collar productivity.
The ServiceNow Reflexivity Problem A crucial irony emerged in the ServiceNow Q3 2026 report. To protect margins, ServiceNow announced a 15% workforce reduction. However, because their Fortune 500 customers were concurrently cutting 15% of their own workforces, those customers canceled 15% of their ServiceNow seat-based licenses. This “reflexive” destruction of the revenue base proved that the AI-driven headcount cuts intended to save firms were mechanically destroying the software sector’s own top line.
The Zendesk Smoking Gun and Private Credit Private credit, which grew to $2.5 trillion, faced its first major default in September 2027 when Zendesk missed covenants on a $5 billion ARR-backed facility. The loan assumed customer service revenue was “recurring,” but AI agents began resolving tickets autonomously without generating tickets at all, rendering the underlying business model obsolete.
Regulatory Arbitrage and the Offshore Web The “Permanent Capital” illusion was unmasked. Firms like Apollo and Athene had built an elaborate offshore architecture in Bermuda and the Cayman Islands. They used offshore SPVs to maximize returns through regulatory arbitrage, investing life insurance policyholder annuities into the very private credit assets that were now defaulting. When these loans were impaired, the spider web of linked balance sheets proved too opaque to untangle in real time, threatening the savings of “Main Street” households.
The Mortgage Question and the Global Collapse Unlike 2008, the 2028 mortgage crisis involved 780+ FICO prime borrowers. These loans became “not money good” because the structural impairment of white-collar income meant borrowers could no longer afford the futures they had borrowed against.
Geographic Concentration: Tech hubs like San Francisco, Seattle, and Austin saw home values fall by 8-11% as the marginal buyer vanished.
India’s IT Collapse: The crisis went global as India’s $200B IT services sector (TCS, Infosys, Wipro) collapsed. The marginal cost of an AI coding agent plummeted below Indian developer wages, leading to an 18% fall in the Rupee and an emergency IMF intervention.
--------------------------------------------------------------------------------
5. Institutional Trajectories: The Three ‘Middle Paths’
The circulation of AI gains depends on Diffusion, Competition, and Institutions. Failure in these areas led to three specific “middle paths”:
The Grace-Period Trap: Slow diffusion due to organizational inertia created a false sense of security. Institutions mistook the absence of immediate catastrophe for safety, wasting the “grace period” needed to build adaptive frameworks.
Democratized Displacement: Open competition distributed AI tools widely, but this created a “race to the bottom.” Firms were forced to automate by competitive pressure—if they didn’t replace humans with agents, their rivals would.
Performative Preparedness: Governments produced politically legible but structurally insufficient responses (e.g., workforce training taskforces). These created an illusion of adequacy while failing to address the fundamental need for fiscal redesign.
The Four Clocks Mismatch: This institutional failure was driven by the differing speeds of complex societies: AI Development (months), Institutional Response (years), Human Adaptation (decades), and Cultural Shift (generations).
--------------------------------------------------------------------------------
6. Cognitive Obstacles to Effective Policy Response
Proactive response was hindered by Identity-Protective Cognition. As research into “Cultural Cognition” shows, individuals process threatening information through a “Stroop-like interference effect,” where identity-threatening data is flagged and dismissed at the earliest stage of cognitive processing, before conscious comprehension begins.
Crucially, higher intelligence did not prevent this. Individuals scoring highest on cognitive reflection tests used their System 2 reasoning to more effectively rationalise group-aligned positions and resist counter-evidence. Analytical skill merely “armed the bias.”
Key Cognitive Traps:
Motivated Reasoning: Cognition as a goal-directed process to protect identity rather than seek accuracy.
The Affect Heuristic: Fast evaluative feelings (safe vs. threatening) substituted for analysis under the cognitive load of the crisis.
The Illusion of Explanatory Depth: The error of believing one understands complex AI macroeconomics better than they do.
Naive Realism: The conviction that one’s own perception of the “productivity dividend” was objective reality, dismissing dissenters as irrational.
--------------------------------------------------------------------------------
7. Proposed Recovery Frameworks and the ‘Transition Economy’
The recovery requires a fundamental repair of the economy’s circular flow.
The Transition Economy Act: Focuses on direct fiscal transfers to displaced workers to replace lost payroll taxes.
The Shared AI Prosperity Act: Proposes a sovereign wealth model, establishing a public claim on the returns of intelligence infrastructure to fund household dividends.
Minimum Viable Recovery Steps:
Stabilise Human Demand: This is the absolute priority. A floor must be built under household income through automatic stabilizers to prevent the deflationary spiral from hardening.
Repair Incentive Structures: Use antitrust and competition policy to break rent-extraction chokepoints and prevent the concentration of AI gains.
Tempo Control: Manage the speed of AI diffusion through liability rules and safety standards to allow institutions time to adapt.
The Canary in the Coal Mine The crisis asks whether our institutions are capable of “checking the ventilation” before structural damage becomes irreversible. The “canary”—the initial sector-specific displacement—gasped for air in 2026. The 2028 post-mortem reveals that while the economy can find a new equilibrium, the disorderly unwind of the human intelligence premium has left the social fabric permanently frayed. The question remains: can we build new frameworks before the next step-function jump renders our current responses obsolete?
