TEXT ANALYSIS: Oracle of Obsolescence

TEXT START: "The real question is not 'Will AI take my job?' but 'How will my job evolve because of AI?'"

1. THE DISSECTION

This is a middle-ground reframe—deliberately calibrated to be comforting without being dismissive. The author, a business school academic, positions himself as the voice of reason between two extremes he characterizes as panicky and dismissive. The argument:

• "Only 25% of job tasks are automatable" (sourced to Goldman Sachs)
• Therefore: no wholesale job destruction
• Therefore: the real problem is organizational lag in redesigning jobs
• Therefore: the responsibility falls on employers to adapt

This is framed as pragmatic wisdom. It is, in fact, a category error dressed in managerial language.

2. THE CORE FALLACY

The central error is treating this as a workforce adaptation problem when it is a structural collapse problem.

The article assumes:
- Jobs are the unit of economic relevance
- Changing 25% of tasks is "manageable disruption"
- Organizational redesign can translate productivity gains into "better jobs"

The DT lens exposes the smuggled premise: That productive participation is the baseline condition. The entire piece conceptualizes human economic value as measured by task execution within job structures. It never questions whether those structures survive the mechanism it ignores.

This is not a middle-ground insight. This is the middle stage of a collapse—the period where the system still looks functional because 75% of the existing structure remains, while the 25% quietly severs the circuit that makes the other 75% economically coherent.

3. HIDDEN ASSUMPTIONS

• Task substitution is the only relevant unit. The author maps AI onto discrete tasks, never asking what happens when AI begins composing tasks—the meta-function of work design itself. He notes AI changes "what I spend my time thinking about." He misses that this is premonitory, not anomalous.

• Organizational inertia is the primary friction. He complains that firms haven't redesigned roles in response to AI. He treats this as failure of will. It is more likely the signal that the firms themselves are not yet sure their current structure has a future.

• "Shadow productivity" indicates adaptive capacity. The author treats informal AI adoption by workers as evidence of workforce readiness. Under DT logic, this is efficiency being extracted from the existing human workforce without creating new human employment. It is not a resilience indicator. It is an extraction pattern.

• "Better jobs" is a viable output. The author assumes redesigned roles with AI-collaboration produce viable economic participation. He never defines "better" in terms of scale. Even assuming every individual job is redesigned optimally, the question remains: will there be enough redesigned jobs to employ the workforce at scale? The math of the thesis doesn't care about role quality.

• Government and employer response is the lever. "The role of government should thus be to enable both skills development and organisational change." Under P3 of DT—Productive Participation Collapse—institutional intervention can delay collapse but cannot reverse the structural mechanism. The author treats institutional responsiveness as the cure. It is, at best, a lag extender.

4. THE SOCIAL FUNCTION

Classification: Ideological Anesthetic — Prestige-Signaling Academic Variant

This article serves a specific class function in the transition period. It:

• Provides plausible deniability for organizational inertia — "it's not that AI is dangerous, it's that managers are behind the curve"
• Offers career-coaching framing ("workers need to develop judgment to evaluate AI outputs") that creates the comforting illusion of actionable agency
• Performs false precision with the 25% figure — giving readers a number to anchor their sense of manageable risk
• Redirects blame away from the structural mechanism and onto employer "unwillingness"

Most critically: It satisfies the institutional audience. It tells firms and governments their role is relevant, their interventions matter, their legacy structures can be adapted. This is the precise narrative that allows decision-makers to manage the transition for the incumbents while the structural math runs its course.

The author is not stupid. He correctly identifies real dynamics. He is analyzing individual task-level mechanics with perfect accuracy while being systemically blind to the logic those mechanics embody.

5. THE VERDICT

The article diagnoses the symptoms of a collapsing circuit as if they were the full clinical picture, then prescribes treatment that only works if the patient's underlying condition is survivable.

The Goldman Sachs 25% figure is not a ceiling. It is a lag indicator—the currently visible boundary of displacement. AI capabilities are not static. The 25% grows. The "cognitive adjunct" function the author describes (AI handled "surrounding work" → author spent more time on "questions that mattered") is precisely the displacement curve. You clear the surrounding work first. Then the adjacent work. Then the "creative" work. Then there is a question left worth asking, and no visible reason a human had to ask it.

The "jobless growth" concern he raises in Singapore is the DT thesis described by its symptoms and misdiagnosed as a redesign failure. It is not a redesign failure. It is the mathematical inevitability of a system where productive participation no longer requires mass human labor.

The responsibility he places on "organisations that employ them"—well-credentialed, technically correct—will not be met at the scale or speed required, because the organizations themselves are operating under the same structural incentive that DT identifies: displace labor or be displaced.

This article is a transitional sedative. It is useful to the extent it helps you understand what the anesthetic class believes, and why they need to believe it.