The Number That Cuts Against the “AI Is Killing Tech Jobs” Story
The dominant narrative of the past two years has been that AI is hollowing out technology employment. The hiring data for the second half of 2026 tells a different story. According to Robert Half’s July 2026 research, 78% of technology leaders plan to increase permanent headcount in the second half of 2026 — a strong majority intending to grow, not shrink, their teams.
That figure sits well above the broader employer market. Across all U.S. employers, the same research found 66% plan to increase permanent hiring in the second half of 2026, compared to 60% in the first half and 57% one year prior — a steady upward march. Technology, at 78%, is running hotter than the economy-wide average. Employers are not retreating from tech talent under pressure from automation; they are leaning in.
The catch is that intending to hire and being able to hire are two different things. And it is the gap between them — not any AI-driven collapse in demand — that defines the 2026 labor market.
The Crunch: Demand Is High, Supply Is the Problem
The constraint is unambiguous in the data. In Robert Half’s technology-specific research, 65% of technology hiring managers say finding skilled talent is more challenging than it was a year ago. Across the wider employer base, 58% report that finding qualified talent is more difficult than a year ago. Either way, the same picture emerges: the appetite to hire is outrunning the available supply of qualified people.
The consequences are already showing up as lost work. In the tech-specific data, 71% of technology leaders report skills shortages caused project delays in the past year, and 49% say projects were canceled entirely. Nearly half of leaders killed planned work not because the business case failed, but because they could not staff it. That is the opposite of an oversupplied labor market.
And the roles driving the crunch are exactly the ones the AI narrative claimed would be automated away. The same research shows AI, ML and data science postings grew 163% from 2024, and security roles rose 124% year-over-year. The demand is concentrated in the highest-skill categories, and it is precisely there that supply is thinnest. The hardest-to-find capabilities employers named were not exotic — the same research listed industry-specific knowledge (47%), software proficiency (42%), and leadership abilities (40%) as the top gaps. The shortage is one of depth and combination, not of raw applicants: employers are drowning in resumes while starving for people who can actually do the specialized work.
Advertisement
AI Skills Are Now the Hardest Hire on Earth
The starkest confirmation comes from outside the US market. ManpowerGroup’s 2026 Global Talent Shortage Survey — covering 39,063 employers across 41 countries — found that for the first time ever, AI skills have surpassed all others to become the single most difficult capability for employers to find worldwide, overtaking traditional engineering and IT.
Within that survey, AI Model & Application Development (20%) and AI Literacy (19%) now lead the global ranking of hard-to-find skills, ahead of engineering, sales, and manufacturing. Notably, this is happening even as overall hiring difficulty eases slightly, with 72% of employers reporting difficulty filling roles, down from 74% the prior year. The broad talent market is loosening; the AI-skills market is tightening. Those two trends moving in opposite directions is the whole story of 2026 hiring.
What This Means for Job-Seekers and Talent-Exporting Markets
A global demand-supply mismatch for verified specialists is not a threat to skilled workers — it is an opening. The practical question is how to position for it.
1. Target the categories where cancellations are happening, not the general market
When 49% of tech leaders cancel projects for lack of staff, those unstaffed projects are the demand you want to meet. Aim your skills at AI/ML, data science, and security — the categories that grew 163% and 124% respectively — rather than the saturated generalist pool. The scarcity is the opportunity, and it is concentrated in named categories.
2. Build for “AI Model & Application Development,” the world’s hardest hire
ManpowerGroup identified AI model and application development as the single hardest skill to find globally. That is an explicit instruction for where a motivated learner should specialize. Depth here — not breadth across many tools — is what employers cannot source and will pay to secure.
3. Make your competence verifiable, because employers are screening harder
With hiring managers reporting it is harder to identify genuinely qualified candidates, the constraint is trust, not just skill. Ship public projects, contribute to real codebases, and produce work an employer can inspect. In a market where demand outruns provable supply, verifiability is the differentiator that moves you to the front of the line.
4. Position for remote and contract demand, not just local permanent roles
Because 56% of employers also expect to hire contract professionals for specialized expertise, the opening is not limited to full-time local roles. English-capable specialists in talent-exporting markets can meet global demand through remote and contract work — reaching the 78% of tech leaders who are hiring regardless of where the candidate sits.
The Structural Lesson
The 2026 hiring data forces a correction to the prevailing story. AI is not eliminating the demand for technology talent; it is raising the bar on what talent must be able to do, and the supply of people who clear that bar is not keeping up. A market where 78% of tech leaders want to grow, half are canceling projects for lack of staff, and AI skills have become the hardest hire on the planet is not a market in retreat — it is a market with a bottleneck. And bottlenecks are where individual leverage lives. For skilled, verified specialists — especially those in markets that can serve global demand remotely — the mismatch between soaring hiring intent and scarce qualified supply is the clearest opportunity the labor market has offered in years. The winners will be the people who close the specific, named skill gaps that employers have already told the world they cannot fill.
Frequently Asked Questions
If AI is automating tech work, why are 78% of tech leaders still hiring?
Because the data points the other way. Robert Half’s July 2026 research found 78% of technology leaders plan to increase permanent headcount in the second half of 2026, running ahead of the 66% figure across all US employers. AI is raising the bar on what talent must be able to do rather than removing the demand for it.
Which specific skills are hardest for employers to fill?
For the first time ever, AI skills have overtaken all others as the single most difficult capability to find worldwide, according to ManpowerGroup’s survey of 39,063 employers across 41 countries — with AI Model and Application Development (20%) and AI Literacy (19%) leading the ranking. On the demand side, AI, ML and data science postings grew 163% from 2024 and security roles rose 124% year-over-year.
What is the shortage actually costing companies?
Unstaffed work. 71% of technology leaders report that skills shortages caused project delays in the past year, and 49% say projects were cancelled entirely. Nearly half of leaders killed planned work because they could not staff it — which is what makes those named scarcity categories the openings worth targeting.
Sources & Further Reading
🔗 Related Intelligence
61% of Tech Leaders Plan to Grow Headcount in H1 2026: Which Roles Are Hiring Now
“The Market Did Not Recover; It Split”: Tech Hiring’s Two-Speed Reality
Tech Hiring Rebounds 21%: Finance Leads, AI Fluency Locks In as Baseline
Skills Gaps Are Killing Tech Projects: 49% Canceled and the Careers That Fix It
