I surveyed workers to see if AI had caused job losses and was surprised by the findings

TEXT START: Do artificial intelligence tools such as ChatGPT eliminate jobs, create new ones, or both?

The Dissection

This is a cautious survey report trying to answer a macroeconomic question with narrow personal testimony. It measures whether currently employed, self-selected panelists believe AI directly caused discrete events. It does not measure task displacement, reduced hiring, wage suppression, hours, nonreplacement, or workers excluded because they are already jobless.

The Core Fallacy

It treats the absence of reported AI-caused job loss as evidence that AI has not substantially altered the labor market. That is the wrong unit of analysis. Automation can eliminate tasks, reduce future openings, and weaken bargaining power without producing immediate mass firings. The survey counts survivors and excludes many of the displaced. Under the Discontinuity Thesis, this is lag evidence—not a refutation.

Hidden Assumptions

• Workers can correctly identify whether AI caused or prevented a job change.
• AI’s effects appear as a new job, promotion, or termination rather than hiring freezes or intensified work.
• Currently employed respondents represent the entire labor force.
• A self-selected online panel reflects typical workers.
• The period since 2023 is long enough to reveal terminal displacement.
• AI-related promotions are durable, despite the text admitting agentic AI may obsolete those skills.

Social Function

Partial truth functioning as ideological anesthetic and transition lullaby. The data credibly shows limited visible, self-attributed disruption so far. Its danger is converting that delay into reassurance while the study’s design cannot detect slower structural decay.

The Verdict

The study does not show that AI failed to destroy jobs. It shows that the corpse count was taken among employed survivors using a test that misses slow suffocation. The 3% figure is evidence about immediate, attributable layoffs—not about task erosion, labor-market entry, or productive participation. In DT terms, this is a snapshot of the lag before social death becomes unmistakable, not a challenge to the discontinuity thesis.