AI jobocalypse: What can be done to address the defining business challenge of our time?

TEXT START: Artificial intelligence (AI)-related job displacement is coming, though no one knows how fast it will proceed, how far it will go and which sectors it will affect most.

The Dissection

This is a structural collapse disguised as a business-management problem. The article concedes that AI will displace workers, then attempts to contain the damage through productivity growth, new startups, retraining, tax incentives, and corporate goodwill.

Its central maneuver is to preserve the comforting sequence of the old order: firms adopt technology, productivity rises, prices fall, demand expands, new jobs appear, and workers transition into them. It treats AI as another general-purpose technology while quietly acknowledging that AI can perform the cognitive tasks that previously generated wages, consumption, and bargaining power.

The article identifies real lag mechanisms: integration friction, uncertain costs, regulatory delays, worker retraining, and temporary human-AI complementarity. But it mistakes these brakes for a reversal of direction. CEO declarations about employee development measure rhetoric, not durable demand for human labor.

The Core Fallacy

The core fallacy is equating increased output and demand with increased employment.

The Jevons effect can expand the market for a product without expanding the need for human workers to produce it. If AI lowers the cost of cognitive production, firms can sell more while requiring fewer people. More startups can mean more automated entities, thinner margins, and fewer human roles—not a restored mass-employment circuit.

The article’s examples work only while humans remain economically necessary. The nurse practitioner assisted by AI is a plausible transitional winner. But if the system eventually diagnoses and treats more cheaply and reliably, augmentation becomes substitution. The same tool that upgrades the worker also maps the worker’s tasks into software.

Under the Discontinuity Thesis, P1 is conceded but underweighted, P2 is almost entirely ignored, and P3 is treated as a temporary skills mismatch. Taxing tokens or subsidizing training may delay deployment. Neither changes the competitive pressure toward superior, cheaper machine labor.

Hidden Assumptions

The article smuggles in several assumptions:

• Firms will continue producing roughly the same goods and services rather than restructuring around AI-native output.
• Expanded demand will remain labor-intensive.
• Human workers can retrain faster than AI capabilities advance.
• There will be enough durable human-specific tasks to absorb displaced workers.
• Corporations can be induced to retain workers without sacrificing competitive position.
• Token usage can be measured and taxed without substitution into foreign providers, open models, local inference, or other inputs.
• More entrepreneurship will translate into broad employment rather than automated oversupply.
• Abundance will be distributed through wages rather than captured by owners of AI capital.

The most important assumption is left unstated: that the wage-to-consumption mechanism can survive after productive participation becomes optional for most people. That is the assumption the thesis rejects.

Social Function

Primary classification: ideological anesthetic and transition management. Secondary classifications: partial truth and elite self-exoneration.

The article offers respectable policy tools for the lag period, but frames a system-level ownership crisis as a solvable question of corporate responsibility and tax calibration. This relocates the burden onto governments to design clever incentives, workers to retrain repeatedly, and corporations to display empathy while competition continues to reward substitution.

Its language of a future of plenty is especially evasive. Material abundance does not restore income, status, autonomy, or productive necessity. UBI and transfers may preserve consumption, but they do not preserve participation. The population can be fed while becoming economically redundant.

The Verdict

This is a polished hospice memo for the employment system. It correctly describes adoption friction and several temporary niches, but it mistakes the delay before displacement for evidence that displacement can be managed away.

The article’s proposals can soften the landing, preserve selected servitor roles, and buy institutions time. They cannot restore the post-WWII mass-employment circuit once AI achieves durable superiority across cognitive work. It offers a brake pedal to a machine whose destination is determined by ownership, cost, and competitive necessity.