---
type: "synthesis"
articles: ["a072", "a084", "a091", "a099", "a102"]
tags: ["human-capital", "judgment", "complementarity", "accountability"]
id: "cross-judgment-accountability"
sources: ["futures"]
sourceVaultSlug: "hbr-seg-futures"
originDay: 2
articleStem: "hbr-seg-futures"
sourceUrl: "(unified vault: 11 sources)"
sourceTitle: "HBR — Futures / Macro — where it's all going"
---
# What Machines Can't Sign

The corpus's humanist through-line: as AI cheapens *production*, the durable premium shifts to *accountable judgment* — and firms are dismantling the very pipelines that produce it.

**A084** is the theoretical core: [[concept-complementarity]] and [[concept-induced-demand]] (from *Prediction Machines*) explain why cheaper code *raises* the value of judgment; [[claim-sign-off-is-product]] / [[contrarian-sign-off-is-product]] insist the liable human sign-off *is* the product; and [[concept-capability-debt-d2]] + [[concept-judgment-debt]] name the invisible liabilities from cutting juniors and de-skilling seniors — a [[concept-tragedy-of-commons-slow-motion]]. The radiology analogy ([[claim-hinton-radiology-error]]) is the parable; the remedy is [[framework-ai-accountability]] and [[action-pair-senior-junior]].

**A099** confirms it from the moat side: [[claim-professional-services-disruption]] erodes the human-capital moat *but* judgment, liability, and trust resist full automation.

**A072** shows the labor-market fallout: [[claim-human-capital-roi]] and [[contrarian-education-roi]] — specialized degrees move from priceable *risk* into unpriceable *uncertainty* — plus [[question-doctor-definition|"what is a doctor in 2035?"]].

**A091** supplies the empirical anecdote: [[claim-genai-lacks-depth]] — GenAI is a "starter pack," impressive but shallow.

**A102** operationalizes the human premium: [[concept-bridger|bridgers]] and [[quote-trust-and-risk|"people don't take risks with those they don't trust"]] — the relational judgment no model can sign for. See [[cross-relocating-scarcity]].