Tech Job Openings Surge Despite AI Automation Concerns

The numbers, taken at face value, tell a reassuring story. Software developer job postings are up 11% year over year [1]. AI and machine learning roles exploded 163% in 2025 [2]. Eighty-seven percent of technology leaders say they plan to increase headcount in the first half of 2026 [3].

But beneath the aggregate figures lies a labor market undergoing its most significant restructuring since the dot-com bust — one where the winners and losers are being sorted not by skill or experience, but by which side of an AI dividing line their job description falls on.

The Numbers Behind the Surge

The headline statistics are real. AI, ML, and data science roles totaled 49,200 postings in 2025, up 163% from the prior year, while cybersecurity roles reached 66,800 postings, a 124% year-over-year increase [2]. Employer job postings mentioning AI jumped 117% between 2024 and 2025 [4]. The Bureau of Labor Statistics projects U.S. computer and IT employment will grow from 6.09 million in 2025 to 7.03 million by 2035 [5].

Yet the picture changes when you look at where these gains are concentrated. The share of AI/ML jobs in the overall tech job market ballooned from 10% in 2023 to 50% in 2025 [6]. That concentration means much of the "tech hiring surge" is really an AI hiring surge, with other categories stagnating or contracting.

Source: Bureau of Labor Statistics (CES5051200001)

Data as of Mar 25, 2026CSV

Federal data tells a more sober story about traditional tech employment. Bureau of Labor Statistics figures show that the computer systems design and related services sector — which covers a broad swath of the technology workforce — employed 344,100 people in February 2026 (preliminary), down from 365,100 a year earlier and 401,500 in February 2024 [7]. The broader information sector has declined from 2.895 million in December 2024 to 2.812 million in February 2026 [7]. These declines are happening even as AI-specific postings spike, confirming that the market is rotating rather than uniformly expanding.

Who's Hiring — and What For

The hiring surge breaks down unevenly by company type. Late-stage companies have increased their hiring rate by 26%, while early-stage startups have seen their rate drop 35% from 49% to 27% over two years [8]. This suggests that mature companies flush with AI investment capital are driving the bulk of demand, while younger firms face tighter funding conditions.

Non-tech industries have become major employers of technical talent. Financial services and manufacturing each approached 100,000 tech role postings in 2025, as organizations invested in securing systems, enabling automation, and managing more complex IT environments [3]. Healthcare, automotive, and government agencies are also investing heavily in technical talent, particularly in AI/ML, cybersecurity, and data engineering [9].

The roles themselves have shifted. Commercial and business-facing technical positions had the highest hiring rate at 34.7%, up 5%, while engineering, marketing, product, and people functions all declined [8]. The most in-demand titles include AI product managers, solutions architects, DevOps engineers, and cloud architects — roles that blend technical ability with business strategy [4].

Big Tech's share of new graduate hiring has dropped sharply. The proportion of new hires at the "Magnificent Seven" (Alphabet, Amazon, Apple, Meta, Microsoft, NVIDIA, and Tesla) who are recent graduates fell from 32% in 2019 to just 7% today [10]. Startups hire 30% faster than large tech firms — 12 days versus 42 on average — but offer roughly 11% lower base salaries [11].

The Entry-Level Collapse

The starkest trend in the data is the disappearance of junior positions. Entry-level tech hiring has dropped 73% in the past year, compared to a 7% overall decline across all job levels [8]. In the UK, tech graduate roles fell 46% in 2024, with projections of a further 53% drop by 2026 [12].

The arithmetic is straightforward: a senior developer using AI coding assistants can now do work that previously required a team that included junior members. As Stack Overflow reported, companies are finding that "a senior developer with an AI assistant is more productive than ever — reducing the need for junior support" [13].

This creates a troubling pipeline problem. The share of juniors and graduates in IT employment has dropped from approximately 15% to 7% over three years [12]. Job postings labeled "entry-level software engineer" grew 47% between October 2023 and November 2024, but actual hiring into those roles dropped 73% in the same period — suggesting companies are posting aspirationally while filling positions with experienced workers [8]. The New York Fed reports that recent graduates face elevated joblessness and over 41% underemployment [12].

The Compensation Divide

Salary trends reflect the bifurcation. The median software engineer salary in the U.S. sits at $130,000 in 2026 [6], with base-pay increases for tech workers expected to rise 3.5% — down from 4% in 2025 [14]. Robert Half data shows even more modest gains of 1.6% for tech salaries broadly [15].

AI specialists, however, occupy a different compensation universe. AI engineer salaries averaged $206,000 in 2025, a $50,000 jump from the prior year [16]. Machine learning engineers lead at $212,928 annually, with computer vision specialists and robotics engineers close behind at $200,000 to $210,000 [16]. Companies pay a 28% salary premium for AI capabilities over traditional tech roles [14].

Source: Rise / Ravio / Robert Half salary data

Data as of Mar 25, 2026CSV

The premiums are even steeper in niche areas. Large language model engineers earn 25% to 40% more than general ML engineers. MLOps specialists who can deploy and scale AI systems command 20% to 35% premiums. AI safety and alignment specialists have seen a 45% salary increase since 2023 [16].

These numbers coexist with overall tech compensation that remains below its 2021-2022 peaks for many roles. The market has not returned to growth-at-all-costs compensation — it has instead concentrated its spending on a narrower set of specialists.

What Happened to AI Job Displacement?

The dire predictions of mass technological unemployment have not materialized — at least not yet. The Information Technology and Innovation Foundation found that AI created approximately 119,900 jobs in 2024, roughly 9.4 times more than the 12,700 jobs it displaced that year [17]. Only 0.1% of all layoffs in 2024 were attributed to AI [17].

However, the trajectory is accelerating. Employers reported about 55,000 layoffs attributed to AI in 2025, representing 4.5% of all job losses — a ninefold increase from the prior year [18]. A March 2026 Duke University CFO survey found that 44% of surveyed CFOs plan AI-related layoffs, projecting cuts of roughly 0.4% of the total U.S. workforce — approximately 502,000 roles [18]. Half of those losses are expected in white-collar positions.

"It's not the doomsday job scenario that you might sometimes see in the headlines," said John Graham, co-author of the Duke CFO Survey, though he acknowledged the pace of displacement is intensifying [18].

The World Economic Forum's Future of Jobs Report 2025 offered a longer-range projection: 92 million jobs displaced by 2030, offset by 170 million new roles created — a net gain of 78 million [19]. Goldman Sachs estimates that AI could displace 6% to 7% of the U.S. workforce if widely adopted, but characterizes the impact as "transitory" as new opportunities emerge [20].

The gap between intent and execution remains large. While 88% of organizations now use AI in at least one business function, only 14% have production-ready deployments, and a mere 11% are actively using agentic AI systems [4]. Goldman Sachs economists found "no meaningful relationship between productivity and AI adoption at the economy-wide level" — a phenomenon that echoes Robert Solow's famous 1987 observation that computers were "everywhere except in the productivity statistics" [18].

Geographic and Sectoral Shifts

The tech hiring surge is reshaping which cities matter. Traditional hubs remain dominant, but secondary markets are growing. Los Angeles ranked 9th nationally with 5,544 tech job postings as of February 2026, while Denver placed 14th with 3,223 postings [21]. Phoenix saw its tech talent pool increase 5.6% between 2021 and 2024 [21].

Remote work availability has contracted. As of Q4 2025, 65% of job postings were fully on-site, 24% hybrid, and just 11% fully remote [22]. This marks a continued pullback from pandemic-era flexibility, though nearly 80% of employees whose jobs can be done remotely are working either hybrid (52%) or fully remote (26%) — suggesting a gap between what companies post and how they actually operate [22].

The sectoral distribution of tech hiring has broadened meaningfully. Financial services, manufacturing, healthcare, and government are all increasing their technical headcount, particularly in cybersecurity and data roles [3]. Over 90% of organizations anticipate IT skills shortages by 2026, with an estimated $5.5 trillion in lost productivity tied to the gap [5].

Historical Parallels — and Their Limits

Previous automation cycles offer partial guidance. When cloud computing scaled between 2008 and 2015, it initially destroyed traditional IT infrastructure jobs — system administrators, on-premises server managers, help desk technicians who maintained physical hardware. But cloud-related job growth started at roughly 6% in 2012, rose to 8% in 2013, and reached 14% by 2015 as the ecosystem matured [23]. The lag between technology adoption and net job creation was approximately three to five years.

The offshoring wave of 2000 to 2005 followed a similar arc. Initial job losses in programming and customer service were eventually offset by growth in project management, systems architecture, and integration roles that required domestic presence and domain expertise.

The ITIF's historical analysis notes that job losses as a percentage of total employment have "trended downward" over 30 years despite continuous automation [17]. An MIT study found that 11.7% of the labor market could "in principle" be automated, though actual displacement has consistently lagged theoretical potential [17].

AI may follow the same pattern, but the timeline is uncertain. Unlike cloud computing, which primarily changed where work was done, AI tools can change what work is done — and by whom. The collapse of entry-level hiring suggests the displacement effects may concentrate at the bottom of the experience ladder rather than across it.

The Sustainability Question

Several indicators suggest the current hiring pattern may not persist. Tech job postings remain 36% below pre-2020 levels according to Indeed's 2025 Tech Talent Report [5]. The 178,000 tech employees laid off in 2025, with projections reaching 211,000 by year's end, show that cuts in non-AI roles are ongoing [5].

The comparison to the 2021 hiring peak is instructive. That cycle was fueled by zero-interest-rate monetary policy and pandemic-driven digital acceleration, producing unsustainable hiring velocity that resulted in the mass layoffs of 2022 and 2023. The current surge is narrower — concentrated in AI, security, and data roles — and coincides with declining employment in traditional tech subsectors.

Some analysts argue companies are hiring now in anticipation that AI tools will mature enough to reduce headcount later — a kind of bridge staffing strategy. The Duke CFO survey data supports this interpretation: companies are investing in AI capabilities while signaling future workforce reductions [18].

Early 2026 saw tech layoffs surpass 45,000 globally, with 68% concentrated in the U.S., as companies restructured to prioritize AI investments [4]. The simultaneous hiring and firing — adding AI specialists while cutting generalists — may be less a paradox than a preview of how workforce transitions actually unfold: messily, unevenly, and with significant pain for those on the wrong side of the shift.

What the Data Doesn't Show

Several important limitations deserve acknowledgment. Job posting data from platforms like Indeed, LinkedIn, and CompTIA can double-count positions listed on multiple sites, inflate apparent demand through "ghost postings" that companies have no immediate intent to fill, and lag actual hiring decisions by weeks or months. The 73% gap between entry-level postings and actual entry-level hires illustrates how misleading raw posting data can be [8].

AI-attributed layoff figures are almost certainly undercounted. Companies rarely cite AI explicitly when reducing headcount, preferring terms like "restructuring" or "efficiency improvements." The true number of jobs displaced or transformed by AI tools is likely higher than any single data source captures.

And the most consequential effects may be ones we cannot yet measure: the roles that were never created, the companies that scaled with ten people instead of fifty, the projects that proceeded without a junior team. These counterfactuals shape the labor market as powerfully as any layoff announcement, but they leave no trace in the data.