---
id: "concept-skill-diversity-reduction"
type: "concept"
source_timestamps: ["¶12", "¶15"]
tags: ["skill-requirements", "labor-demand", "automation"]
related: ["concept-ai-automation-displacement", "claim-skill-requirement-shifts", "action-reskill-automation-roles"]
definition: "The shrinking number of skills required by employers in job postings for automation-prone roles, indicating a hollowing out of the occupation's complexity."
sources: ["reskilling"]
sourceVaultSlug: "hbr-seg-reskilling"
originDay: 10
articleStem: "hbr-edu-35-ai-changing-labor-market"
sourceUrl: "https://hbr.org/2026/03/research-how-ai-is-changing-the-labor-market"
sourceTitle: "Research: How AI Is Changing the Labor Market"
---
# Skill Diversity Reduction

**Definition:** The shrinking number of skills required by employers in job postings for automation-prone roles, indicating a hollowing out of the occupation's complexity.

Skill diversity reduction is a phenomenon observed in job postings for **automation-prone occupations**. As generative AI becomes capable of handling the structured and repetitive tasks associated with these roles, employers list **fewer required skills** in their job descriptions. The research registered a **7% decrease** in required skills for these roles, alongside a **drop in the emergence of new skills**. This reduction signals a hollowing out of the role's complexity, leaving workers highly vulnerable to displacement unless they develop non-automatable skills.

This is the causal mechanism behind [[concept-ai-automation-displacement]] and the demand-side evidence underpinning [[claim-skill-requirement-shifts]]. Because it exposes workers to displacement, it is the trigger for [[action-reskill-automation-roles]] and is voiced urgently by Srinivasan in [[quote-retraining-essential]].

**Enrichment / confidence note:** The working paper ([[entity-displacement-or-complementarity-paper]]) explicitly reports falling skill requirements in automation-prone jobs and rising skill complexity in augmentation-prone jobs. The **−7% figure** is an article-specific statistic, not a widely replicated parameter — treat it as an internal research estimate rather than a consensus benchmark. Yale's Budget Lab ([[evidence-yale-budget-lab]]) does not directly measure posting-level skill diversity and finds no major composition shift so far, a useful caution.