AI Job Postings Are Up 88 Percent — and Most of the Demand Is Outside the Tech Sector

Amid a year of AI layoff headlines, one dataset keeps moving the other way. Dice’s September 2026 Jobs Report, released this month using Lightcast data on more than seven million US tech postings, counts AI and machine learning job postings up 88 percent year on year — roughly five times the 17 percent growth rate for tech postings overall. Technology led all industries in month-on-month posting growth at 30 percent, with finance and banking up 26 percent, healthcare up 23 percent and aerospace and defence up 22 percent as Workforce Observer summarised.

🔍 THE BOTTOM LINE

The AI hiring market is real, measurable, and — the detail most people miss — mostly not happening inside technology companies. More than half of postings asking for AI skills come from outside the tech sector, which means the career door is open to accountants, clinicians, manufacturers and public servants, not only engineers. For New Zealand readers, the same pattern shows up in local data: SEEK counted AI skill mentions in a record 4.2 percent of NZ job ads in September, up 88 percent year on year, almost double the Australian rate — a coincidence of numbers worth double the scepticism, but a consistent direction of travel.

What the fastest-growing titles actually are

The report’s month-on-month title growth list is notable for what is not on it: research scientists. The titles growing fastest — above 150 percent month on month — are Security Architects, Infrastructure Security Engineers, Automation Controls Engineers, Deployment Engineers, Quality Assurance Specialists, Technical Support Analysts, Functional Analysts, Reporting Analysts, Principal Program Managers and Professional Services Consultants.

Read as a group, they describe the work of putting AI into production rather than inventing it: designing and securing the systems, building and testing them, then supporting, reporting on and managing delivery once they are in use. This matches what this site has tracked across the deployment-economy story all year — the forward-deployed engineer counted from 200 postings to 7,300 in three years, and the AI training gig economy paying experts up to US$400 an hour. The bottleneck companies keep reporting is not model capability. It is people who can wire AI into an organisation safely.

Skills-level data says the same thing from the other direction: the year-over-year growth list is headed by Agentic AI, AI Agents, Responsible AI and AI Infrastructure — plus, tellingly, Electronic Engineering and Autodesk Revit, the non-software skills tied to record data-centre construction.

The pay question: a premium with an asterisk

The hiring wave comes with wage data attached, though it is less tidy than the postings curve. Lightcast postings data presented at Randstad’s salary webinar shows jobs advertising at least one AI skill carrying salaries 28 percent above comparable roles with none — roughly US$18,000 a year — and the premium climbing to 43 percent when a posting names two or more AI skills, as HRD reported.

The asterisk: formal pay bands are largely not budging. HRD’s reporting describes employers routing much of the premium through bonuses and performance pools instead of base salary — a two-tier system where an AI-fluent hire and a conventional one can sit on the same band with very different realised pay. For job-seekers the practical read is to negotiate the total package, and to note that the premium is highest precisely where skills are specific to a role rather than generic. PwC’s 2026 global AI Jobs Barometer, analysing more than a billion job ads across 27 countries, puts the average wage premium for workers with AI skills at 62 percent, up from 57 percent the prior year, per PwC’s release — and, as covered in September’s NZ barometer analysis, it likewise finds AI-skilled workers seeing faster wage growth than their industry.

🔍 THE BOTTOM LINE: Two datasets, two magnitudes, one direction: advertised pay for AI-skilled work is rising faster than the market around it. Whether it arrives as base, bonus or contract rate depends on the employer — which is why the growing titles above (consulting, program management, professional services) matter: they are the roles where premiums are being negotiated openly.

The New Zealand read

The Dice report is US-only, so the honest framing is analogue, not home data. But NZ’s own meters have been pointing the same way all spring. SEEK’s September Employment Report found overall job ads up 7.8 percent year on year with AI skill mentions at a record 4.2 percent of postings — 88 percent above last year, and nearly double Australia’s 2.4 percent. Our report on the September data noted the same structural detail Dice’s figures show: ads for the most automatable roles are simultaneously shrinking, meaning the market is rewarding the AI-fluent while quietly repricing the task-list role.

PwC’s NZ barometer counted AI-related job ads rising from about 3,900 to 9,600 in a year. And the skills mix travelling across the Pacific looks like Dice’s: NZ employers are not mostly hiring model researchers, they are seeking AI literacy layered onto existing professional roles — consistent with the small-country pattern where adoption outruns local model-building by design.

How to use this data in a career move

Three practical reads from the numbers:

Sector choice matters more than job title. Finance, healthcare and manufacturing are the fastest-growing AI posting categories after tech itself. An AI-fluent accountant in a bank is in a stronger market than a generic “AI engineer” chasing the same five logo companies.
The deploy roles hire from adjacent skills. Lightcast’s own analysis of the FDE boom found the role shares roughly half its skills with ordinary software engineering; the Dice title list extends that logic to non-engineers — analysts, QA specialists and support engineers are on the fastest-growth list precisely because someone must certify and maintain what the models produce.
Mind the pay structure. With base bands lagging while premiums grow, the negotiation point is variable pay, title and skills certification — and the 43 percent premium for dual-AI-skill roles suggests depth in two AI skills beats breadth in one.

Caveats worth holding

Job postings measure advertised demand, not filled jobs; a posting surge can partially reflect slow hiring cycles rather than pure growth. Dice’s figure is postings pulled on 5 October against a 5 September 2025 baseline, sourced from Lightcast’s commercial database rather than a government survey — it reads employer intent, not payroll outcomes. And a year-over-year comparison against a weak 2025 tech market flatters the percentage (the same reason the 2024 baseline made this comparison look even starker). None of that changes the direction; it only shrinks the decimal.

The other caveat is temporal: the postings boom and the layoff headlines are not contradictions so much as two ends of one reshuffle. Firms are cutting some functions — Barclays’ 5,000 announced cuts being the most-cited example — while building deployment and governance teams at rates the posting data has never shown before. The net employment effect is the open question this site’s NBER churn coverage tracked: surveys so far find most firms reporting no net workforce change, with the mix shifting inside the firm rather than heads leaving it.

FAQ

Are AI job postings really up 88 percent? In Dice’s September 2026 Jobs Report, released October and built on Lightcast data covering more than 7 million US tech postings, AI and machine learning postings grew 88 percent year on year, versus 17 percent for tech postings overall.

Do you need to work in tech to benefit? Per the same data, over half of postings requesting AI skills are from companies outside the technology sector — finance, healthcare and manufacturing are the fastest-growing poster categories after tech itself.

What are the fastest-growing AI-adjacent job titles? Security Architects, Infrastructure Security Engineers, Automation Controls Engineers, Deployment Engineers, QA Specialists, Technical Support Analysts, Reporting Analysts and Professional Services Consultants — production and support roles, not research roles.

How much more do AI skills pay? Lightcast postings data shows a 28 percent advertised salary premium for a named AI skill, rising to 43 percent for two or more — though employers increasingly route the uplift through bonuses rather than base bands.