Jobs of the future: what the 2025 forecasts say and how to test a direction

Where lists of jobs of the future come from

Search for jobs of the future and you get lists of titles, rarely with a word about where they came from. The most quoted list has a clear origin. The World Economic Forum’s Future of Jobs Report 2025 rests on a survey of 1,043 employers with more than 14.1 million employees in 22 industry clusters and 55 economies. The answers were collected from May to September 2024 and describe what those companies expect for 2025 to 2030.

The headline figures need their base stated. Combining employers’ expectations with International Labour Organization employment data, the report estimates that 170 million jobs will be created and 92 million displaced by 2030: a churn of 22% of the 1.2 billion formal jobs in its dataset and a net gain of 78 million, or 7%. The new jobs and the lost ones are rarely the same jobs, which is why skills matter more than the total.

Fastest-growing and fastest-declining roles in the 2025 report

In percentage terms the fastest-growing roles are technological. The list is led by big data specialists, fintech engineers, AI and machine learning specialists and software and applications developers. Data analysts, UI and UX designers and green roles such as renewable energy engineers are in the same top 15. Employers tie this growth mainly to AI, robotics and broadening digital access.

The declining side is mostly clerical: cashiers and ticket clerks, administrative assistants, data entry clerks, bank tellers and postal service clerks. Graphic designers appear just outside the top 10 fastest-declining roles for the first time. The 2023 edition treated them as a moderately growing job, and the authors suggest the change may reflect the growing capacity of generative AI to perform knowledge work.

Absolute numbers tell another story. Farmworkers top the list of the largest growing roles, with 35 million more jobs expected by 2030 in an occupation that already employs over 200 million people. Delivery drivers, construction workers, shop salespersons, nurses, social workers, care aides and teachers follow. Software developers and delivery drivers are the only two roles that appear in both rankings, the fastest-growing and the largest-growing.

Reading a career forecast: growth rate against head count

A percentage in these rankings is growth relative to the number of people now in the role. A small profession that adds a modest number of people can top the chart, while a huge one adds millions and looks unremarkable. Ask both questions about any role: how fast it grows, and how many openings that means in the place where you will look for work.

Then look at who answered. The survey targeted companies with 500 employees or more, the report says that small enterprises and the informal sector are out of its scope, and its authors warn that the conclusions should not be treated as comprehensive. Russia is not among the 55 economies, although Kazakhstan and Uzbekistan are. Employers described expectations within their own organisations, not hiring promises, and the report’s disclaimer says its data is subject to change without notice.

What generative AI changes inside a job, according to the ILO

The International Labour Organization looked at AI from the other end: tasks, not titles. Its working paper of May 2025, written with researchers from Poland’s NASK, scores occupations for exposure to generative AI. One in four workers worldwide is in an occupation with some exposure, and 3.3% of global employment falls into the highest category. Exposure reaches 34% of employment in high-income countries against 11% in low-income ones, and clerical occupations remain the most exposed.

The authors stress that exposure does not imply the immediate automation of an entire occupation. Most occupations consist of tasks that need human input, so the most likely effect is that jobs change; how much depends on how employers adopt the technology and whether workers get the chance to learn it. The World Economic Forum survey shows the same shift at the level of tasks: employers say 47% of tasks are now done mainly by people, 22% mainly by technology and 30% by both, and expect a nearly even split by 2030.

The skills employers name more often than job titles

The more useful part of the report is about skills. Employers expect 39% of workers’ core skills to change by 2030, down from 44% in the 2023 edition. The fastest-growing skills are AI and big data, networks and cybersecurity, and technological literacy, followed by creative thinking, resilience, flexibility and agility, and curiosity and lifelong learning. The top 10 also includes leadership, talent management, analytical thinking and environmental stewardship.

Analytical thinking remains the most common core skill: seven out of ten companies consider it essential. Programming and technological literacy are among the skills that most clearly separate growing jobs from declining ones. AI shows up in hiring plans too: two-thirds of employers plan to hire people with specific AI skills, while 40% expect to reduce staff where AI can automate tasks.

Languages appear as multilingualism and sit in the lower part of the list of rising skills, so their value depends on the market you aim at. The scale of retraining is the figure to remember: if the world’s workforce were 100 people, 59 would need training by 2030, and 11 of them would be unlikely to get it. A title can drop out of a list between two editions; a skill moves with you to the next role.

Russia’s Ministry of Labour forecast of workforce needs

With no Russian sample in the global survey, the national forecast is the better guide for readers in Russia. In January 2025 the Minister of Labour, Anton Kotyakov, presented a five-year forecast to the Federation Council, built on a survey of 260,000 companies with 22 million employees. By 2030, compared with 2022, the economy needs an extra 3.1 million people in employment, against the 2.4 million mentioned earlier.

The picture differs from the global one. Most people are needed in manufacturing, transport and storage, hospitality, professional activities and science, and almost every production industry needs more skilled workers and IT specialists. Needs grow most for vehicle mechanics, builders, finishing workers, electricians and toolmakers, and among highly qualified specialists for software developers, doctors, engineers, database and network specialists, and architects. Two thirds of the demand is for people with secondary vocational education.

In December 2025 the government’s portal reported the ministry’s first seven-year forecast: 12.2 million people to bring into the economy, counting new jobs, the replacement of retirees and development plans. The ministry plans to update the forecast every year, and the minister himself has called it a roadmap rather than a solution to every employment problem.

Choosing a career direction and testing it cheaply

A forecast cannot choose for you, but it can narrow the search. Start from three facts about yourself: the work you can do for hours without forcing it, the market and language you will work in, and the time you can give each week. Cross them with the skills every source above names: data and AI, technological literacy, analytical thinking. A bookkeeper who likes order may move towards data analysis, a salesperson who writes well towards search marketing.

Test the direction before paying for anything long. A free introductory course shows whether you enjoy the hard part, not only the idea. A small project on a real problem, such as automating a weekly report, shows whether you can finish. A conversation with two or three practitioners shows what the job looks like on an ordinary Tuesday. Give each test a week or two and write down what surprised you.

From a one-week test to a programme at VITON13 SCHOOL

Some of this you can practise alone this week. Do one task from your job with an AI assistant, then check every fact it gave you against the source. Write a short script that renames files or counts rows in a table. If you have a website, open it in a search engine’s webmaster tools and see which pages are indexed.

A structured course adds what self-study usually lacks: an order of topics, feedback and a finished piece of work. In the paid programmes of VITON13 SCHOOL practice runs through the programme and continues after it, the final applied assignment comes with feedback, and every learner builds a project connected to the course. The AI course covers working with AI deliberately, prompting, research and verification, automation, AI agents, your own systems and safety. The SEO course covers technical SEO, intent, semantics, architecture, content, schema.org, linking, analytics and indexing in Google, Yandex and Bing, and AI search.

The cheapest test is the free programming course: Python, JavaScript and TypeScript, 16 lessons each, with code that runs in the browser and is checked at once, no registration needed. A week with it tells you more about your interest in programming than another list of jobs of the future. Each paid programme lists its current terms on its own page; compare them with the hours you wrote down and pick one skill to start.