AI Is Going the Wrong Way. A Nobel Economist Explains Why. | 01 - Vocal Media

TEXT START: Daron Acemoglu says we’re building AI to replace people when we should be building it to make people more capable.

The Dissection

This text is a policy intervention disguised as a diagnosis. It correctly identifies the incentives pushing firms toward labor substitution: tax asymmetry, measurable cost savings, and finance departments that reward immediate headcount reduction. It also recognizes that displacement is fast while reinstatement is slow, uncertain, and organizationally expensive.

Its real objective is to preserve labor’s income and institutional role by steering AI toward augmentation. The article does not defeat automation. It argues that society should voluntarily choose a less immediately profitable deployment path.

The Core Fallacy

The article treats AI’s direction as primarily a policy choice after assuming away the competitive mechanism that determines deployment. Under the Discontinuity Thesis, once AI delivers durable cost and performance superiority, firms that refuse substitution lose to firms that adopt it. Tax changes can alter margins; they cannot permanently erase the competitive advantage of cheaper machine labor.

The article’s own distinction between augmentation and substitution does not save it. If augmentation doubles a worker’s output and demand does not expand proportionally, the firm still needs fewer workers. “Pro-worker” AI can therefore become a more efficient instrument of displacement.

The ATM example is a historical niche, not a general law. Reinstatement occurs only when new human tasks are numerous, economically necessary, difficult to automate, and supported by expanding demand. It can create pockets of employment without restoring the mass employment–wage–consumption circuit.

The article correctly describes transition friction. It mistakes that friction for a viable escape route.

Hidden Assumptions

• Demand will expand enough to absorb workers made more productive.
• New tasks will emerge at sufficient scale and remain human-essential.
• Firms can be induced to sacrifice cost minimization for augmentation indefinitely.
• Governments can coordinate taxes, antitrust, procurement, and labor rules against competitive pressure.
• Redistributive institutions will remain competent and politically reliable.
• Workers can retrain faster than displacement proceeds.
• Human skills will remain scarce enough to command wages after AI makes them widely accessible.
• Current enterprise failures reflect implementation problems rather than an early stage of a stronger automation curve.
• Better services and lower prices will compensate workers for losing their production-based claim on income.

Social Function

Primary classification: partial truth serving as transition management and ideological anesthetic, with heavy prestige signaling.

The article provides a useful account of tax distortions, coordination costs, weak current productivity conversion, and the asymmetry between displacement and reinstatement. But it channels those truths into the comforting conclusion that better design can preserve mass participation. It gives policymakers and labor institutions a humane vocabulary for managing decline without confronting ownership of the productive system.

The “Nobel economist” framing also converts a structural threat into a respectable policy debate: adjust incentives, fund training, improve governance, and the old labor bargain may survive. That is institutional reassurance, not a solution to P1–P3.

The Verdict

Acemoglu is right about the proximate mechanics and wrong about their terminal significance. Pro-worker AI can improve public services, reduce transition damage, and preserve selected servitor niches. It cannot guarantee mass productive participation once AI becomes cheaper and more capable across cognitive work and human institutions cannot maintain human-only economic domains at scale.

This is not a refutation of the Discontinuity Thesis. It is a lucid description of the lag phase—and a hospice protocol for labor’s bargaining power dressed as a technology design brief.