Photo by Vitaly Gariev on Unsplash
- Draup analyzed 2.85 million job descriptions from June 2025 to June 2026, finding over 40,000 active postings each for software engineering, data engineering, and DevOps — roles widely predicted to shrink.
- PwC's study of over 1 billion job ads across 27 countries found "professionalised" roles (human expertise layered with AI) growing twice as fast and commanding 42% faster salary growth than "democratised" roles where AI simply lowers the skill floor.
- AI skills now carry a 62% average wage premium as of June 2026 — ranging from 118% in consumer markets down to just 16% in government roles.
- Entry-level positions with AI exposure are 7 times more likely to demand senior-level skills like strategic thinking and leadership from day one, compressing what was once a multi-year apprenticeship into an immediate expectation.
What We Found
2.85 million. That is the number of job descriptions Draup — an enterprise talent intelligence firm — fed through its analysis engine between June 2025 and June 2026. The question they were trying to answer: which skills are companies actually hiring for, versus which ones sound good in earnings calls? The findings, reported by Business Insider and widely circulated across Google News on July 4, 2026, cut against the loudest version of the AI-displacement narrative. As of mid-2026, software engineering, data engineering, and DevOps each had over 40,000 active job postings. Not historical postings — active listings. According to Vijay Swaminathan, Draup's CEO, "AI isn't reducing the need for technical talent, but it is changing what makes technical talent valuable." Most people are sitting with the first clause and skipping the second.
The Evidence
The Draup dataset is one half of a two-source picture worth reading together. On the other side: PwC's 2026 Global AI Jobs Barometer, which analyzed over 1 billion job advertisements across 27 countries and introduced a framework that makes the market's underlying logic legible. PwC divides roles into two categories. "Democratised" roles are those where AI has lowered the skill floor — the task becomes easier for everyone, which also means it becomes cheaper for employers to fill. "Professionalised" roles require human expertise layered on top of AI fluency. As of PwC's June 2026 report, professionalised roles are growing twice as fast and commanding 42% faster salary growth than their democratised counterparts.
Euronews Business's coverage of the same PwC dataset highlighted a striking company-level finding: firms in the top quintile of AI exposure achieved 163% labor productivity growth relative to 2018 and 52% headcount growth — compared to just 36% among the least AI-exposed firms. The companies most exposed to AI are not shrinking their workforces. They are growing faster. According to Joe Atkinson, PwC's Global Chief AI Officer, "The companies seeing the greatest returns on AI are using it to amplify human expertise, accelerate innovation and create entirely new sources of value."
Bloomberg's reporting pinpointed where cuts are actually landing: financial services and information sectors averaged 28,000 job losses monthly in 2026, based on U.S. government employment data. But the headline misses texture. The same companies that eliminated routine positions in early 2026 began rehiring — specifically for customer service and content moderation — in March 2026, after discovering that AI replacements failed in ways they had not anticipated. Employers, Bloomberg noted, are "recalibrating" their understanding of where humans add irreplaceable value versus where automation holds.
Draup's data gets specific about which tools are reshaping job requirements. GitHub Copilot, Cursor, and Claude appeared in over 60,000 job listings across nine job categories analyzed between June 2025 and June 2026. If you want to understand how these tools differ in practice, AI Tools' breakdown of ChatGPT vs. Claude vs. Jasper is worth a read — but in hiring terms, what matters is not which tool wins a benchmark. It is that employers now expect candidates to work alongside them fluently and exercise judgment about when outputs go wrong.
Photo by Vitaly Gariev on Unsplash
What It Means for Your Career Leverage
The wage data is where the market is signaling most clearly. As of PwC's June 2026 report, AI skills command a 62% average wage premium — up from 57% in 2025. But that 62% average obscures a spectrum that shapes every personal finance and salary decision differently depending on your sector.
Chart: AI skills wage premium by sector category, based on PwC's 2026 Global AI Jobs Barometer analysis of over 1 billion job advertisements across 27 countries.
The 118% premium in consumer markets — retail, media, e-commerce — reflects how directly AI fluency translates to measurable revenue outcomes. Government roles at 16% reflect procurement timelines and political constraints on rapid automation. For workers in finance, healthcare, or technology, the relevant number sits between those poles, but the direction is consistent: the premium rose 5 percentage points in a single year, and AI-related job postings grew 69% annually against 9% overall market growth as of PwC's June 2026 analysis — nearly eight times faster.
The entry-level data is the signal I find most clarifying. Roles at the entry level with AI exposure are now 7 times more likely to require senior-level skills — leadership, strategic thinking, accountability — from day one. These seniorised entry positions have grown 35% since 2019, while other entry-level roles declined 10%. The traditional apprenticeship model — spend two years doing routine work, gradually earn more responsibility — is disappearing at the bottom of the funnel. Employers are collapsing that runway. In my read, this is why the market feels contradictory to so many people entering the workforce: low unemployment headlines coexist with genuine difficulty because the entry paths that existed five years ago have structurally narrowed. The World Economic Forum projects 92 million roles will be displaced by 2030 with 170 million new roles emerging — a net gain of 78 million positions — but with 22% workforce churn that is concentrated disproportionately right now, at the routine-task level.
How to Act on This
Before your next performance review or job search, be honest about whether your current role is professionalised or democratised. A useful diagnostic: could a non-expert use an AI tool to produce most of what you produce in a typical day? If yes, you are in the democratised category — wage growth will be slower and job security thinner over the next three years. If your work requires judgment calls that depend on institutional knowledge, client relationships, or regulatory accountability — the kind of accountability where someone named and specific owns the outcome — you are in the professionalised category. That is where careers need to migrate, and where financial planning assumptions about income growth should be anchored.
PwC's data gives workers a concrete leverage point that most are not using. If you have demonstrated fluency with GitHub Copilot, Cursor, Claude, or domain-specific AI tooling, the market is willing to pay a 62% average wage premium for that combination — and most workers are not pricing it into their ask. When a hiring manager raises compensation expectations, here is a script worth practicing out loud: "Based on current market data for roles that require both [your domain] expertise and hands-on AI fluency — which this role clearly does — the range I'm targeting is [X to Y]. I'm glad to walk through specifically how I've been using [tool name] to [concrete outcome]." Specific tool plus specific outcome beats vague "AI experience" every time. Know your BATNA (best alternative to a negotiated agreement) before the conversation — if you have another offer or a clear market rate from PwC's barometer, say so plainly.
Consumer markets already price in the AI premium at 118% — meaning much of that advantage is already competed away for new entrants. The more durable play for personal finance and career ROI is building AI fluency inside domains where adoption is still early and the premium is still forming: healthcare, legal, education, industrial operations. AI-related postings growing at 69% annually means the supply of credentialed workers is not keeping pace with demand. That gap is where upskilling investment — courses, certifications, side projects with real AI tools — compounds most reliably over the next five years.
Frequently Asked Questions
What jobs will AI not replace, and how do I know if mine is at risk?
Roles requiring complex human judgment, client accountability, regulatory interpretation, and cross-functional leadership are the most durable against automation as of mid-2026. Draup's analysis of 2.85 million job descriptions identified the most in-demand skills clustering around judgment, design, and accountability — functions where an AI error carries real-world consequences and someone specific must own the outcome. If your role involves those elements, it sits in PwC's "professionalised" category, which is growing twice as fast as the average. If your work is primarily executing well-defined, repeatable tasks, the risk is higher and the signal to diversify is worth taking seriously now.
How much does having AI skills increase salary in 2026 — is it worth the investment?
As of PwC's June 2026 Global AI Jobs Barometer, the average wage premium for AI skills is 62%, up from 57% in 2025. That figure covers an enormous range — 118% in consumer markets, 16% in government roles — so the most useful number for your personal finance decisions is the sector-specific premium, not the overall average. For most workers in private-sector roles, a 30% to 80% premium for demonstrated AI fluency is a reasonable planning assumption. The investment in learning a specific AI tool that your target employers explicitly list in job postings typically has a faster payback than general AI literacy courses.
Will AI replace software engineers, or are those roles actually growing?
As of mid-2026, Draup's dataset showed over 40,000 active software engineering postings, and GitHub Copilot, Cursor, and Claude appeared in more than 60,000 listings across nine job categories. The picture is not replacement — it is bifurcation. Software engineers who exercise judgment over AI-generated code, design systems, and own architectural accountability sit firmly in the professionalised category PwC identifies as growing twice as fast. Engineers whose primary function is routine code generation are more exposed. The skill that matters most right now is not the ability to generate code with an AI tool — it is the ability to evaluate when that output is wrong and why.
Disclaimer: This article is for informational and educational purposes only and does not constitute financial, legal, or career advice. Individual circumstances vary — consult qualified professionals before making major career or financial decisions. Research based on publicly available sources current as of July 4, 2026.