The AI hiring crisis: Why entry level jobs are disappearing first - BusinessToday

URL SCAN: The AI hiring crisis: Why entry level jobs are disappearing first - BusinessToday
FIRST LINE: Speaking in a McKinsey interview, Brynjolfsson said AI is changing the nature of work by automating specific tasks rather than replacing entire occupations.

The Dissection

The text performs a controlled downgrade of an economic rupture into a workforce-adjustment story. It concedes that AI is eliminating junior work, cites a 16–17% employment decline among exposed workers aged 22–26, and admits that the corporate hierarchy is becoming “diamond-shaped.” Then it redirects the reader toward adaptation, training, creativity, leadership, and managing AI agents.

Its real function is to make the disappearance of the entry point sound like redesign. The article identifies the first visible fracture but refuses to follow it through the system: if firms automate the routine work through which novices acquire experience, they also destroy the pipeline that produces future specialists and managers.

The Core Fallacy

The central error is treating task automation as evidence that occupations will survive. Under the Discontinuity Thesis, occupations do not need to vanish in one administrative event. They can be hollowed out task by task until the remaining human work supports far fewer people.

“Higher-value work,” creativity, judgment, and leadership are not an automatic refuge. They are scarce functions, and AI increases the output of the workers or owners who control it. The result is not one human per redesigned job. It is fewer humans supervising larger AI-augmented production systems.

The proposed solution—learn new skills and direct teams of AI agents—also assumes that companies will continue hiring large numbers of humans to perform that direction. That is precisely the assumption the employment data is attacking. The shrinking entry level is not merely a training problem. It is the first-stage collapse of productive participation under P1 and P3.

Hidden Assumptions

• That AI will remain primarily complementary rather than becoming durably cheaper and better across cognitive tasks.
• That every displaced junior worker can move into a higher-value role instead of competing for a much smaller number of positions.
• That education and retraining can manufacture enough demand to replace the jobs being automated.
• That companies will preserve human career ladders for institutional continuity, even when competitive pressure rewards labor reduction.
• That future managers can be produced without the junior work through which managers traditionally gain experience.
• That wages remaining stable means the damage is limited, when delayed compensation pressure is entirely compatible with a lagged employment collapse.
• That productivity gains will be distributed through new products, better service, and broader opportunity rather than captured by firms and AI-capital owners.
• That “amplifying intention” changes ownership and bargaining power. It does not. A more powerful tool still benefits whoever controls the tool, the data, the infrastructure, and the organization.

Social Function

Classification: partial truth, transition management, and ideological anesthetic.

The partial truth is real: AI is automating tasks, entry-level workers are being hit first, and organizational structures may become less pyramid-shaped. The anesthetic is the claim that initiative, creativity, and training can broadly absorb the displaced. That converts a structural demand problem into an individual failure of adaptation.

The article also provides elite self-exoneration. Executives are encouraged to measure AI by business outcomes rather than labor cost, but no mechanism is supplied that forces those outcomes to generate mass employment. It offers a humane rationale for AI deployment while leaving the ownership structure untouched.

The Verdict

This is an accurate early-warning report wrapped in a false recovery narrative. Entry-level workers are not simply losing obsolete tasks; they are losing the economic doorway through which the system reproduced skilled labor and middle management.

The “diamond-shaped” company is not a stable new equilibrium. It is a narrower funnel: fewer entrants, fewer protected career paths, and greater output concentrated in AI-capital owners and indispensable operators. Training may create individual Servitors. It cannot restore the mass employment–wage–consumption circuit once cognitive labor becomes competitively automatable. The article sees the first corpse and recommends résumé improvement.