"Do You Really Need to Go to College in the AI Era?": A Father's Realistic Advice to His High ...

TEXT START: My son, who is a high school sophomore in the United States, often asks me, "Do I have to go to college? If so, where should I go?"

The Dissection

This article is a collapse memo disguised as career advice. It correctly detects that narrow specialization is losing value and that AI is compressing entire departments into smaller groups of AI-enabled workers. It then redirects that structural fact into an individual obligation: become versatile, logical, AI-fluent, and capable of doing everything.

The real function is to preserve faith in personal maneuverability. College is reframed as a networking and sorting venue rather than an educational necessity. Humanities and philosophy are presented as the alleged human refuge, while companies are urged to preserve entry-level hiring so future judgment can be cultivated.

The most revealing sentence is the claim that workers must become people who can “do everything using AI.” That is not liberation. It is labor compression: one worker is expected to absorb the output of several former specialists while competing against every other worker equipped with the same tools.

The Core Fallacy

The article mistakes residual human involvement for durable human labor demand.

Under the Discontinuity Thesis, the relevant question is not whether humans will still perform judgment, networking, or oversight. Some will. The question is whether those activities will remain necessary at sufficient scale to sustain the wage-and-consumption circuit. This text never confronts that question.

Its errors are mechanical:

• It assumes “what to execute and why” is a permanent human sanctuary. Under P1, planning, judgment, evaluation, and strategic framing are cognitive functions subject to automation. Human judgment may survive as a temporary lag defense or elite preference, not as a mass employment base.

• It treats productivity gains as evidence that work remains. The IT example proves only that fewer people can produce more. In a competitive system, that usually reduces labor demand rather than preserving headcount.

• It mistakes generalist versatility for scarcity. If the same AI systems give everyone cross-domain capability, “AI fluency” becomes a baseline expectation, not a moat. The worker becomes a cheaper universal interface to capital.

• It confuses a company’s need for a small future leadership pipeline with a social need for mass employment. Firms may retain apprenticeships for indispensable Servitors while eliminating most entry-level roles. That does not rescue the majority.

• It assumes coordination can remain human-controlled. Under P2, institutions cannot preserve stable human-only economic domains at scale once AI is cheaper, faster, and strategically superior.

The article sees tasks being automated but refuses to follow the chain to P3: productive participation collapses, ownership captures the gains, and individual adaptability becomes a rationing mechanism for the few positions left.

Hidden Assumptions

• AI remains merely a tool rather than capital controlled by Sovereigns who determine access, hiring, and distribution.
• Human judgment is irreducible and cannot itself be modeled, delegated, or automated.
• Higher individual productivity creates more jobs instead of allowing firms to eliminate workers.
• Employers will sacrifice short-term cost savings to preserve entry-level training.
• College networks remain broadly accessible and produce bargaining power rather than just credentialed competition.
• A person’s ability to use AI will remain differentiated after AI use becomes universal.
• Educational choice can materially alter a macroeconomic process driven by ownership and competition.
• The existence of a few roles requiring human trust or judgment implies a viable labor market for the many.
• The author’s own non-graduation success is transferable rather than a survivorship-biased example.
• Philosophy, humanities, or astrophysics automatically translate into economic indispensability. They do not.

Social Function

This is a partial truth serving as transition management and ideological anesthetic.

Its accurate observations make the sedative effective: specialization is weakening, AI competence matters, and companies that eliminate all junior workers may damage their own succession pipeline. But the article converts systemic displacement into a personal development problem. If someone fails, the implied explanation is insufficient versatility or insufficient judgment—not the ownership structure that captures the productivity gains.

It also provides elite self-exoneration. Companies are asked to behave farsightedly and preserve a thin layer of talent development, while the underlying power arrangement remains untouched. No serious mechanism for distributing AI ownership, income, or productive control is offered.

The Verdict

The article is lucid about the first wave and blind about the destination. College may still function as a network, credential, or sorting mechanism for future Servitors. Philosophy and the sciences may sharpen judgment for a minority who become indispensable to Sovereigns. They do not restore mass productive participation.

“Become someone who can do everything with AI” is not a durable survival doctrine. It is the labor market’s demand that each remaining worker carry more of the corpse. The text correctly senses that the old career ladder is dying, then recommends polishing the rungs as they fall.