---
id: "lib-2026-06-15-2026-global-ai-jobs-barometer"
type: "library"
title: "2026 Global AI Jobs Barometer"
lang: "EN"
org: "PwC"
source_type: "Report"
date: "2026-06-15"
week: "2026-W25"
url: "https://nosov.ai/library/lib-2026-06-15-2026-global-ai-jobs-barometer/"
original_url: "https://www.pwc.com/gx/en/issues/artificial-intelligence/job-barometer/2026/2026-global-ai-jobs-barometer-full-report.pdf"
tier: "S"
tier_name: "Shapes the field"
tier_reason: "PwC's annual flagship on AI and labour markets, built on over one billion job ads across 27 markets plus firm accounts for 52,274 companies; one of the most cited datasets in the AI-and-jobs debate."
summary_basis: "full-text"
original_minutes: "45"
tags: ["jobs", "skills", "productivity", "entry level", "two track labor market"]
frames: ["solow-paradox"]
---

# 2026 Global AI Jobs Barometer

Tier S (Shapes the field): PwC's annual flagship on AI and labour markets, built on over one billion job ads across 27 markets plus firm accounts for 52,274 companies; one of the most cited datasets in the AI-and-jobs debate.

Summary based on the full text.

## In short

PwC's 2026 Global AI Jobs Barometer, "Two futures for jobs in an AI era", analyses over one billion job ads from six continents along with company accounts. Its main finding is a two-track labour market. In some jobs AI takes over the basic tasks and leaves more expert work to people ("professionalised"). In others it takes over the expert tasks and lowers the skill bar ("democratised"). The first group grows faster and sees faster pay growth. More AI-exposed companies also show faster growth in productivity, headcount and wages.

## Key numbers

- 22% vs 52%: share of advertised jobs AI is professionalising (more expert) vs democratising (less expert)
- 39% vs 17%: job-posting growth since 2018, professionalised vs democratised jobs
- 35%: growth since 2019 in AI-exposed US entry-level roles that now demand senior skills; other such roles fell 10%
- 62%: advertised wage premium for roles requiring AI skills, up from 57% the year before

## What it reports

- 22% of advertised jobs are professionalised, 52% democratised and 26% have low AI exposure. Of 380 ISCO occupations, 74 are professionalised, 125 democratised and 181 low-exposure.
- Since 2018, professionalised job postings grew 39% and democratised 17%. Skills demanded grew 68% vs 33%.
- Since 2021, advertised salaries rose 37% for professionalised jobs and 26% for democratised jobs, which PwC reports as "42% faster" growth.
- Skills for the most AI-exposed jobs are changing 2.2x faster than for the least exposed, a gap 75% wider than last year. New tasks in exposed roles are 2.5x more likely to rely on empathy, judgment or creativity (the MIT EPOCH measure).
- Among US entry-level jobs, 52% of new skills in the most exposed roles are "traditionally senior", against 7% in the least exposed. "Seniorised" entry roles grew 35% from 2019 to 2025, and non-seniorised ones fell 10%.
- Productivity (revenue per employee) grew 33.5% since 2018 in the most AI-exposed companies, against 24.0% in the least exposed. The top 20% of the exposed group reached 163%.
- Headcount grew 52.2% vs 35.7% and wages 24.4% vs 16.6% (most vs least exposed). Only 8% of CEOs say AI raised revenue more than slightly in the past year.
- AI specialist hiring grew eight times faster than all hiring in 2025. Tech, media and telecoms leads at 11.4% of postings.

## How they got it

PwC UK economists combined Lightcast job-posting data for 27 markets with a refreshed version of Felten's AI occupational exposure index. Occupations were classed as professionalised or democratised using expertise-change scores from Teeselink and Carey (2026). The company results use ORBIS financial data for firms with at least $50m in turnover, with each firm's AI exposure assigned from its sector.

## Limits

- Company-level "exposure" is a sector label, not measured AI use. The firm sample excludes companies that exited the market (survivorship bias), and PwC says the results are not causal.
- "42% faster" is a relative figure: the absolute gap is 11 points (37% vs 26%). The introduction gives the exposed-firm productivity figure as 34%, the body as 33.5%.
- Entry-level and seniorisation results rest mostly on US postings (about 73% of the long-run sample). Ads show what employers ask for, not who gets hired.

## Why it matters for companies

The useful question for any role is whether AI is taking the routine part or the expert part. The answer shapes hiring, pay and training differently. Junior roles exposed to AI now ask for stakeholder management and decision-making, so onboarding and mentoring need to build those skills in the first year. The headcount results do not show that AI creates jobs. They show that firms in AI-exposed sectors grew while adopting it, so companies should test their own cases instead of assuming cuts or growth.

Original: https://www.pwc.com/gx/en/issues/artificial-intelligence/job-barometer/2026/2026-global-ai-jobs-barometer-full-report.pdf