What happens to the talent pipeline if AI takes all the entry-level jobs? - The Globe and Mail

TEXT START: The entry-level job appears to be an endangered species.

The Dissection

The article correctly identifies the first fracture: AI is removing the low-level tasks through which institutions trained future professionals. But it frames this as a talent-pipeline management problem—an invitation to redesign training, teach soft skills, and reinvest savings.

That diagnosis stops one layer short of the machinery. The pipeline is not merely changing. Its economic purpose is being destroyed. The entry-level worker was both labor and trainee; AI removes the labor first, then makes the training expense harder to justify. The managerial ladder becomes a narrow funnel, not a ladder available to the mass of entrants.

The Core Fallacy

The article assumes that eliminating transactional work will reliably move humans upward into judgment, creativity, communication, and oversight. Under the Discontinuity Thesis, that is the central error.

P1 does not stop at spreadsheets, clerical work, or data collection. AI advances into analysis, drafting, checking, communication, and eventually much of the judgment built from those functions. “Soft skills” are not automatically durable economic moats; they become another layer of cognition available for automation or supplied by a much smaller number of workers.

The article also assumes that firms will use AI gains to train surplus humans. Firms generally use automation to reduce labor cost and concentrate output. Without ownership or coercive redistribution, productivity gains do not create a guaranteed human employment ladder.

Hidden Assumptions

• Future managers and specialists will be needed in roughly the same numbers as before.
• AI supervision will require a large human workforce rather than a small verification elite.
• Employers will absorb the cost of training workers whose initial tasks no longer produce immediate value.
• Fiduciary duty and professional trust will preserve broad accounting employment, rather than concentrating it among a smaller set of highly trusted operators.
• Side hustles, content creation, and portfolio work can replace structured apprenticeships at meaningful scale.
• Human error in AI outputs will remain frequent and costly enough to support mass human review.
• The ownership question is irrelevant: who controls the AI capital is treated as less important than who acquires the next credential.

These assumptions convert a distributional crisis into a curriculum problem.

Social Function

Primarily transition management and ideological anesthetic, with a substantial partial truth component.

The article is useful because it exposes the apprenticeship collapse and the coming shortage of experienced human operators. But its proposed remedy—upskilling, soft skills, reinvestment—allows institutions to discuss labor displacement without discussing ownership, bargaining power, or the shrinking number of economically necessary humans.

“AI is a productivity tool, not a replacement” is not analysis. It is a temporary description of the present used as a promise about the future. The “universe shifting” language performs the same function: it turns forced exclusion into a natural evolution workers are expected to admire while adapting to it.

The Verdict

The article sees the smoke but mistakes the fire for a pipeline blockage. The real mechanism is productive participation collapse: AI removes the entry rung, then erodes the higher rungs built on experience acquired there. P1 automates cognitive work; P2 prevents institutions from preserving stable human-only domains; P3 leaves a minority of owners, controllers, trusted verifiers, and indispensable maintainers economically necessary.

The future talent pipeline will not primarily produce a larger class of managers. It will sort a smaller number of humans into Sovereigns, high-trust Servitors, and a large surplus population competing for shrinking transition niches. Retraining may create exceptions. It will not restore the mass employment circuit.