Artificial intelligence is already rewriting job descriptions, salaries, and entire career paths. Drawing on research from the World Economic Forum, McKinsey, PwC, Goldman Sachs, and Stanford University, here is what the next decade of work actually looks like — and how to stay ahead of it.
Picture a hospital ward in 2035. A radiologist still reads the scan — but an AI system has already flagged the shadow on the X-ray, compared it against millions of similar cases, and drafted the first version of the report. The radiologist’s job has not vanished. It has changed completely.
That story is already playing out across banks, factories, law firms, and marketing departments, and it is moving faster than most people realize. According to the World Economic Forum’s Future of Jobs Report 2025, the next wave of technological, economic, and demographic change will create 170 million new jobs by 2030 while displacing 92 million others — a net gain of 78 million jobs worldwide, but a churn equal to 22% of all employment on the planet.
So will artificial intelligence steal your job, change it, or hand you a better one? The honest answer: it depends on which job you do, which skills you build, and which decisions you make between now and 2035. Here is what the evidence actually shows.
Source: World Economic Forum, Future of Jobs Report 2025 — survey of 1,000+ employers representing 14M+ workers across 55 economies.
01 / Where things standAI’s Grip on the Workplace, Right Now
AI is no longer a futuristic idea confined to research labs. It already sits inside the daily operations of banks, hospitals, factories, and marketing teams.
PwC’s 2026 Global AI Jobs Barometer, which analyzed more than one billion job ads across 27 countries, found that companies most exposed to AI grew headcount by 52% since 2018, compared with just 36% growth at the least-exposed companies. In other words, heavier AI adoption currently correlates with more hiring, not less, at the company level.
Wages tell a similar story. Workers with verified AI skills now earn a 62% wage premium over peers in similar roles without those skills, up from 57% the year before and just 25% two years before that, according to PwC. That premium climbs as high as 118% in consumer-facing industries.
However, the picture is not uniformly rosy. Goldman Sachs estimates that roughly 300 million full-time jobs globally are exposed to some degree of AI automation, with about two-thirds of current occupations in the US and Europe touched by the technology in some way. Most of that exposure is partial — AI handles a slice of a job, not the whole thing — but for around 7% of roles, Goldman’s analysts believe the work could eventually be fully automated.
Meanwhile, McKinsey Global Institute calculates that AI-powered agents and robots could generate roughly $2.9 trillion in US economic value every year by 2030, once businesses redesign workflows around people, software agents, and physical robots working as a team, rather than automating one task at a time.
Put simply: AI is already reshaping paychecks, hiring plans, and job titles in 2026. The next decade will only accelerate that shift.
“This research shows that the power of AI to deliver for businesses is already being realised.”— Carol Stubbings, Global Chief Commercial Officer, PwC
02 / The numbers, comparedHow Many Jobs Will AI Really Affect?
Ask five different research organizations how many jobs AI will affect, and you’ll get five different numbers. That isn’t because anyone is wrong — each report measures something slightly different: exposure, automation potential, augmentation, or net change.
Here is how the major studies line up against one another:
| Organization | Report | Headline finding |
|---|---|---|
| World Economic Forum | Future of Jobs Report 2025 | 170M jobs created, 92M displaced — net +78M by 2030 |
| McKinsey Global Institute | Generative AI & the Future of Work | Up to 30% of US work hours automatable by 2030; $2.9T/yr in value |
| Goldman Sachs | How AI Will Affect the Labor Market | 300M jobs globally exposed; two-thirds of US/EU jobs touched |
| PwC | 2026 Global AI Jobs Barometer | AI-exposed firms grow headcount 52% vs 36%; 62% wage premium |
| Stanford HAI / Digital Economy Lab | AI Index Report 2026 | Entry-level developer jobs (ages 22–25) down ~20% since 2024 |
Despite the different numbers, three things show up in nearly every report. First, the overall effect on total employment is likely to be positive, or close to neutral — AI is expected to create roughly as many jobs as it removes, if not more. Second, the disruption is uneven: it concentrates on routine, codified, desk-based tasks rather than physical or judgment-heavy work. Third, the transition will not feel calm. Even reports projecting net job growth, like the WEF’s, describe churn equal to a fifth of the entire global labor market within five years.
Source: PwC, 2025 & 2026 Global AI Jobs Barometer — wage premium for workers with AI skills over peers in similar roles without them.
“The combination of significant labor cost savings, new job creation, and higher productivity for non-displaced workers raises the possibility of a productivity boom that raises economic growth substantially.”— Jan Hatzius, Chief Economist, Goldman Sachs
03 / Sector by sectorIndustry by Industry: Who Adapts, Who Shrinks
AI does not affect every industry the same way. Some sectors are being quietly automated from the back office outward. Others are simply getting faster, smarter versions of the same human-led work. Here is how the picture looks across seven major industries.
Healthcare: Augmented, Not Replaced
AI is becoming a second pair of eyes for clinicians, not a replacement for them. Diagnostic tools now flag abnormalities in scans, predict patient deterioration before symptoms appear, and draft clinical notes automatically. McKinsey highlights a pharmaceutical company that used generative AI to draft clinical study reports, cutting review time by nearly 60% and reducing errors by about half. Even so, healthcare remains one of the fastest-growing employment categories on Earth, driven by aging populations rather than automation. The real bottleneck isn’t robots replacing nurses — it’s too little caregiving capacity for too many patients.
Finance: From Number-Crunching to Judgment Calls
Financial services sits near the top of every AI-exposure list, and for good reason. Algorithms already handle fraud detection, basic underwriting, and a growing share of routine compliance checks. JPMorgan Chase’s CEO Jamie Dimon said in 2025 that the bank had already automated roughly a fifth of certain back-office functions using AI. Yet PwC’s data shows financial services among the biggest beneficiaries of AI-driven productivity, with revenue per employee rising three times faster in AI-exposed industries than in less-exposed ones. Expect fewer junior analysts doing pure data entry by 2035, but rising demand for professionals who can interpret AI output and advise on decisions a machine shouldn’t make alone.
Manufacturing & Logistics: Robots Handle Repetition, Humans Handle the Unexpected
Factory floors have used robotics for decades, but AI is now making those robots adaptive instead of fixed. PwC’s 2026 Barometer found that manufacturing now posts a higher share of job ads requiring AI skills than even financial services, a sign of how aggressively the sector is investing. Logistics is shifting too, as route-optimization software and early autonomous-vehicle trials reshape long-haul trucking and delivery work. Still, unpredictable, hands-on tasks — emergency repairs, custom installations, quality judgment calls — remain stubbornly hard to automate, which is why skilled trades keep growing even inside heavily automated industries.
Retail & Customer Service: The Most Visible Front Line
This is where AI’s impact is most visible to ordinary consumers. Chatbots resolve a large share of routine queries, self-checkout has spread widely across US retail, and AI scheduling tools are replacing manual rostering. McKinsey’s modeling found that demand for clerks, cashiers, and administrative assistants could fall by millions of positions in the US by 2030, since these roles are dominated by repetitive, data-heavy tasks. The US Bureau of Labor Statistics projects cashier roles will shrink by more than 300,000 positions by 2034, the largest absolute decline of any occupation it tracks.
Creative Industries & Media: Augmentation With an Asterisk
Generative AI can now draft marketing copy, translate documents, and produce rough video edits in seconds. Yet the WEF’s research also lists creative thinking as one of the fastest-rising core skills employers want, precisely because AI-generated content still needs human taste and brand judgment to be useful. The likely split by 2035: AI absorbs the rough first draft and templated content, while human creatives move upstream into strategy, concept, and quality control.
Software & Technology: The Disruptor Gets Disrupted
Few industries have embraced AI faster than the one building it. Coding assistants have helped developers complete routine tasks more than 50% faster, and Stanford researchers found that AI tools which solved only 4.4% of real-world software-engineering problems in 2023 could solve 71.7% of similar problems by 2024. That leap explains an uncomfortable fact: employment for software developers aged 22 to 25 has fallen by nearly 20% since 2024, even as senior engineers remain in high demand. The skill that matters now isn’t writing code line by line — it’s architecture, debugging AI-generated code, and product judgment.
Education: The Personal Tutor at Scale
AI tutoring tools can adapt explanations to a student’s pace in real time, something no single teacher can do for 30 students at once. The WEF projects continued growth in education roles globally, driven by expanding student populations in lower-income regions and rising demand for lifelong learning everywhere else. Rather than replacing teachers, AI is shifting their role toward mentorship and the social, motivational coaching that software still cannot replicate.
04 / The other half of the storyThe New Jobs AI Is Creating
While headlines focus on disappearing jobs, a quieter and faster story is unfolding: entirely new job titles are appearing on hiring platforms that simply did not exist five years ago.
Industry hiring-platform data shows AI engineer roles growing roughly 143% year over year through 2025–2026, prompt-engineer postings rising about 136%, and AI content-creator roles climbing nearly 135%. A widely cited 2017 Dell Technologies and Institute for the Future study, still referenced across today’s workforce research, estimated that as much as 85% of jobs that will exist by 2030 have not been invented yet.
| Role | What they actually do | Typical salary (USD) |
|---|---|---|
| AI Engineer | Builds and deploys AI models inside real products | $150K–$250K+ |
| Prompt Engineer | Designs reliable instructions and workflows for AI systems | $96K–$170K |
| AI Ethicist / Governance Lead | Sets policy on fairness, bias, and safe AI use | $93K–$170K |
| AI Security Analyst | Defends AI systems against manipulation and data poisoning | $150K–$210K |
| Chief AI Officer | Leads company-wide AI strategy at the executive level | $200K+ |
| AI/Human Workflow Specialist | Redesigns processes so people and AI agents work together | $85K–$140K |
| Context / RAG Engineer | Connects AI models to trustworthy company data sources | $110K–$180K |
These roles cluster around three needs that barely existed before generative AI went mainstream: building AI systems, governing them responsibly, and integrating them into existing human workflows. The more interesting story by 2035 may not be which jobs AI takes away, but which categories of work it makes possible for the first time.
05 / What employers wantThe Skills That Will Matter Most by 2035
If job titles are changing this fast, skills matter more than degrees. The WEF found that 39% of core skills required across the global workforce will change by 2030, and that figure climbs even faster inside AI-exposed roles, where PwC found required skills shifting 66% faster than in less-exposed jobs.
Two categories of skill are rising together, not competing with each other:
Technical skills on the rise
- AI & big data literacy
- Cybersecurity & networks
- General technological literacy
- Programming & automation tools
Human skills on the rise
- Analytical & critical thinking
- Resilience, flexibility & agility
- Creative thinking
- Leadership & social influence
This pairing matters. The single most in-demand skill in the WEF’s survey is not technical at all: analytical thinking, which seven in ten employers call essential. The reason is simple — AI can generate options and analysis quickly, but someone still has to judge whether the output is correct, relevant, and fair.
PwC’s entry-level data tells the same story from another angle. Entry-level jobs most exposed to AI are now seven times more likely to require traditionally senior skills, such as independent judgment and leadership, than entry-level jobs in less-exposed fields. AI hasn’t lowered the bar for getting hired — in many cases, it has raised it, even at the very bottom of the career ladder.
By 2035, four forces converge on the workplace at once: industries restructure, new AI-native jobs emerge, demanded skills shift, and governments race to write the rules. None of these forces moves alone.
06 / The generational divideWhy Young Workers Are Feeling It First
If AI’s impact still feels abstract, ask someone who graduated in the last two years.
A landmark Stanford University study, led by economist Erik Brynjolfsson and based on payroll data from millions of American workers, found a 13% relative decline in employment among 22-to-25-year-olds in occupations most exposed to generative AI, including software development, accounting, and customer service. Employment for experienced workers in those same fields kept growing throughout the same period.
By 2026, the gap had widened further. Entry-level software-developer employment for workers under 25 had fallen by nearly 20% since 2024, according to Stanford’s AI Index Report. Crucially, researchers found this pattern only in jobs where AI automates tasks outright. In jobs where AI mainly augments human work — helping someone do their job better, rather than doing it for them — youth employment held steady or even grew.
The explanation researchers offer is almost poetic: AI is very good at replacing “book learning,” the codified knowledge fresh graduates bring straight from the classroom. It is far worse at replacing the tacit, hard-won judgment that comes only from years on the job. That’s uncomfortable news for new graduates — but it also points to the fix: pairing formal education with real, supervised experience as early as possible, rather than waiting for a “safe” moment that won’t arrive on its own.
07 / The human moatThe Jobs Least Likely to Disappear by 2035
Not every job sits on the same side of this story. Across nearly every major study, the same categories keep showing up as the most resistant to automation.
| Category | Example roles | Why AI struggles here |
|---|---|---|
| Healthcare & caregiving | Nurses, therapists, home health aides | Empathy, physical touch, real-time judgment in emergencies |
| Skilled trades | Electricians, plumbers, HVAC techs | Every job site differs; almost nothing is documented the same way twice |
| Emergency response | Firefighters, paramedics, EMTs | Split-second decisions in unpredictable, high-stakes settings |
| Leadership & people management | Executives, HR leaders | Vision, trust, and conflict resolution are relationship-based, not data-based |
| Skilled creative direction | Creative directors, architects | Combines originality with accountability for the final call |
| Law & ethics-heavy work | Judges, senior legal counsel | Carries legal and moral accountability that can’t be outsourced to software |
The common thread is accountability paired with unpredictability. A plumber tracing a hidden leak relies on instinct built from hundreds of unrepeatable jobs. A judge’s ruling carries legal weight a model cannot hold. As one widely cited workforce analysis puts it, the safest jobs of the next decade need “physical presence, emotional intelligence, creative judgment, or moral reasoning” — four qualities AI still cannot convincingly fake.
None of this makes these jobs AI-proof forever. It makes them AI-resistant for now — and that resistance is exactly why they are also among the fastest-growing categories in nearly every major jobs report.
08 / What to do about itPreparing for 2035: A Practical Roadmap
The data is clear on one point above all others: standing still is the riskiest career move available right now. Here’s what the evidence suggests for three different groups.
For workers
Treat AI fluency as a baseline skill, not a specialty. The WEF’s Reskilling Revolution initiative aims to give 1 billion people better skills and economic opportunity by 2030 — but no government program can replace personal initiative. Coursera reported that generative-AI course enrollments jumped from roughly two per minute in 2023 to six per minute in 2024. Start with the AI tools used in your own field, then build toward judgment-heavy skills — leadership, ethics, complex problem-solving — that remain hard to automate.
For businesses
PwC’s research is blunt about this: the companies winning right now aren’t simply buying AI software. They are redesigning workflows so people and AI agents work as a team, with humans handling exceptions, ethics, and relationships while AI handles volume and speed. Firms that treat AI purely as a cost-cutting tool tend to shrink; firms that treat it as a growth engine tend to hire faster than their competitors.
For students and early-career workers
Favor roles and internships involving real client or patient contact over purely back-office, repetitive tasks, since those are precisely the tasks most exposed to automation. Build a portfolio of judgment calls you’ve made, not just tasks you’ve completed — that’s evidence neither employers nor AI systems can manufacture for you.
09 / Beyond the numbersInequality, Policy, and the Choices Ahead
A figure like “78 million net new jobs” can sound reassuring, until you remember it’s a global average, not a personal guarantee. The person who loses a job to automation is rarely the same person who gets hired into the new AI-related role next door.
This is why economists increasingly frame AI’s labor impact as a policy challenge as much as a technological one. The WEF notes explicitly that whether this transition produces shared prosperity or deeper inequality depends substantially on policy choices made in the 2025–2035 period, not on the technology alone.
Some of that policy groundwork is already visible. The European Union’s AI Act is setting transparency and safety standards that other regions are likely to copy, while national reskilling programs and “future of work” task forces have spread across dozens of countries since 2018. Anthropic, the AI company, has separately published research exploring policy responses to AI’s economic effects — a sign that even AI developers increasingly recognize that adoption without adaptation creates real social risk.
“AI is the most profound technology humanity is ever working on.”— Sundar Pichai, CEO, Google & Alphabet, on the scale of disruption ahead
Pichai has been candid about this tension, acknowledging that society “will have to work through societal disruption” as AI reshapes jobs once considered untouchable — including, in his words, his own. That kind of honesty from industry leaders matters, because this transition will not manage itself.
FAQFrequently Asked Questions
Will AI take my job by 2035?
It depends heavily on what your job involves. Roles built around repetitive, codified tasks — data entry, basic bookkeeping, routine customer support — face the highest risk. Roles built around physical unpredictability, emotional connection, or moral accountability, such as nursing, skilled trades, and leadership, are far more resistant. Most major studies, including the WEF’s, project net positive job growth overall — but that masks real disruption at the level of individual roles.
What jobs will AI replace first?
Data-entry clerks, telemarketers, basic bookkeepers, cashiers, and routine paralegal or coding tasks consistently rank as highest-risk across McKinsey, Goldman Sachs, and Stanford research, mainly because these roles involve structured, repeatable work that generative AI already handles well.
What new jobs will AI create?
Roles like AI engineer, prompt engineer, AI ethicist, AI security analyst, and human-AI workflow specialist barely existed in their current form before 2023. Several are now growing more than 100% year over year, according to multiple hiring-platform analyses.
Will AI create more jobs than it destroys?
Most major institutional forecasts say yes, narrowly. The WEF projects a net gain of 78 million jobs by 2030. However, “net positive” hides real pain for individuals whose specific roles disappear, even when the aggregate numbers look healthy.
How can I future-proof my career against AI?
Build AI fluency inside your current field, develop the skills AI still struggles with — judgment, leadership, hands-on problem-solving, emotional intelligence — and seek roles involving direct client, patient, or community contact rather than purely back-office processing.
Which industries are safest from AI disruption?
Healthcare delivery, skilled trades, emergency services, and senior leadership roles consistently rank as the most resistant to automation, mainly because they combine physical unpredictability with high accountability — two things AI still struggles to replicate.
The bottom lineWhat Happens Next Is Still Up to Us
Nobody can tell you with certainty what your job will look like in 2035. What the evidence does show, consistently and from very different sources, is the shape of the transition: routine and codified work will keep shrinking, judgment-heavy and relationship-heavy work will keep growing, and the gap between workers who adapt early and those who wait will keep widening.
That is, in a sense, good news. Unlike a sudden economic shock, this transition is visible, well-documented, and still unfolding gradually enough to act on. The World Economic Forum frames the years ahead as “the Intelligent Age” — and intelligence, both human and artificial, will define who thrives in it.
The workers, businesses, and policymakers who treat the next decade as a planning window, rather than a countdown to disaster, are the ones most likely to look back on 2035 and see opportunity rather than upheaval. The technology is moving fast. How we respond to it is still entirely up to us.
Sources & Further Reading
- World Economic Forum — Future of Jobs Report 2025
- McKinsey Global Institute — Generative AI and the Future of Work in America; AI: Work Partnerships Between People, Agents, and Robots
- Goldman Sachs Research — How Will AI Affect the US Labor Market?
- PwC — 2026 Global AI Jobs Barometer
- Stanford Digital Economy Lab / Stanford HAI — Canaries in the Coal Mine? AI Index Report 2026
- US Bureau of Labor Statistics — Occupational Employment Projections, 2024–2034