--- title: "AI Replacing Jobs Statistics: Forecasts vs Data | Nexum" description: "AI replacing jobs statistics, sourced and dated: BLS and World Economic Forum projections next to what Dallas Fed and MIT Sloan measured." url: https://nexumautomations.com/statistics/ai-job-displacement updated: 2026-10-31 publisher: Nexum Automations, an AI consulting firm --- > Nexum Automations is an AI consulting firm that builds what it recommends. It finds where AI pays off in a company, builds it on the tools the team already uses and stays until it is live and working. Statistics # AI replacing jobs statistics, forecast and measured AI replacing jobs statistics, sorted by what they show: projections from the BLS and the World Economic Forum, and observed change from the Dallas Fed and MIT Sloan. Last reviewed October 30, 2026 6 min read The numbers ## Key numbers on AI job displacement, with sources * 92M jobs projected to be displaced by 2030, about 8% of current employment, in a global employer survey. [World Economic Forum, Future of Jobs Report, 2025 ↗](https://www.weforum.org/publications/the-future-of-jobs-report-2025) * 170M new jobs projected to be created, equal to 14% of today's total employment. [World Economic Forum, Future of Jobs Report, 2025 ↗](https://www.weforum.org/publications/the-future-of-jobs-report-2025) * 78M net jobs added in the same projection, a 7% rise in total employment. [World Economic Forum, Future of Jobs Report, 2025 ↗](https://www.weforum.org/publications/the-future-of-jobs-report-2025) * \-1% change in employment since late 2022 in the 10% of sectors most exposed to AI. [Federal Reserve Bank of Dallas, research note, 2026 ↗](https://www.dallasfed.org/research/economics/2026/0224) * +3% growth in employment share over five years for high-wage roles heavily exposed to AI. [MIT Sloan, AI and the US labor market, 2023 ↗](https://mitsloan.mit.edu/ideas-made-to-matter/how-artificial-intelligence-impacts-us-labor-market) * \-3.5% fall in employment within firms for the top-paying roles over five years. [MIT Sloan, AI and the US labor market, 2023 ↗](https://mitsloan.mit.edu/ideas-made-to-matter/how-artificial-intelligence-impacts-us-labor-market) * 6% higher employment growth over five years linked to a large increase in AI use at a firm. [MIT Sloan, AI and the US labor market, 2023 ↗](https://mitsloan.mit.edu/ideas-made-to-matter/how-artificial-intelligence-impacts-us-labor-market) * +35% projected growth for data scientists, 2025 to 2035. [U.S. Bureau of Labor Statistics, Occupational Outlook Handbook, 2025 ↗](https://www.bls.gov/ooh/math/data-scientists.htm) * +10% projected growth for software developers, QA analysts and testers, 2025 to 2035. [U.S. Bureau of Labor Statistics, Occupational Outlook Handbook, 2025 ↗](https://www.bls.gov/ooh/computer-and-information-technology/software-developers.htm) * +5% projected growth for lawyers, 2025 to 2035, and the same for accountants and auditors. [U.S. Bureau of Labor Statistics, Occupational Outlook Handbook, 2025 ↗](https://www.bls.gov/ooh/legal/lawyers.htm) * 0% little or no change projected for paralegals and legal assistants, with technology including AI limiting demand. [U.S. Bureau of Labor Statistics, Occupational Outlook Handbook, 2025 ↗](https://www.bls.gov/ooh/legal/paralegals-and-legal-assistants.htm) * +2% projected growth for real estate brokers and sales agents, slower than average. [U.S. Bureau of Labor Statistics, Occupational Outlook Handbook, 2025 ↗](https://www.bls.gov/ooh/sales/real-estate-brokers-and-sales-agents.htm) * 6% of small-business workers using AI say they use it to automate workflows with little human involvement. [U.S. Chamber of Commerce Foundation, Main Street AI Monitor, 2026 ↗](https://www.uschamberfoundation.org/workforce/half-of-small-business-workers-use-ai-most-to-boost-productivity-not-automate-jobs) Our data ## What buyers see in this search The first page of Google for "ai replacing jobs statistics", counted by type of result. | | Results | | ---------------------------------------------------- | -------------- | | Articles by companies that sell a product or service | 4 of 9 | | Official, academic or standards sources | 2 of 9 | | Forums and Q&A threads | 1 of 9 | | Social and blog-platform posts | 1 of 9 | | Independent publishers and explainers | 1 of 9 | | Titles that answer this exact search | 1 of 9 | | Google features on the page | An AI Overview | The 9 organic results on the first page of Google, United States, desktop, measured with DataForSEO on October 29, 2026\. Types only: no company or product is named. ## Projected versus measured Projected * Forecasts for future years * Net gain in jobs overall * Built from surveys and models Measured * Changes already recorded * Small drop in exposed sectors * Task shifts inside jobs AI replacing jobs statistics come in three kinds. Exposure counts tasks that software could do. A projection estimates future employment. Observed change counts what already happened. Projections point to churn more than collapse. The World Economic Forum expects 170 million jobs created and 92 million displaced by 2030, a net gain. That is a forecast from employers, not a count. Measured data is thinner. The Dallas Fed finds employment down 1% since late 2022 in the sectors most exposed to AI. MIT Sloan finds the effect lands on tasks inside jobs more than on whole occupations. ## Job displacement due to AI statistics, by role * Software developers Projected growth of 10%. * Data scientists Projected growth of 35%. * Lawyers, accountants Projected growth of 5% each. * Paralegals Little or no change projected. * Real estate agents Projected growth of 2%. The BLS projects each occupation separately, so the picture differs by role. The guides below read each one in depth. Growth is projected for [software engineers](https://nexumautomations.com/guides/will-ai-replace-software-engineers) at 10% and for [data analysts](https://nexumautomations.com/guides/will-ai-replace-data-analysts), where data scientists are projected up 35%. [Lawyers](https://nexumautomations.com/guides/will-ai-replace-lawyers) and accountants are each projected up 5%, and the BLS says automation of routine work is not expected to cut overall demand. [Paralegals](https://nexumautomations.com/guides/will-ai-replace-paralegals) are the exception, with little or no change projected and AI named as a brake on demand. For real estate agents the projection is 2%. ## How to read AI job loss statistics Check first * Who was measured * Which year * Exposure, projection or observed Watch out for * Forecasts told as facts * Mixed populations * Rounded-up percentages Start with who was measured: all US workers, one occupation, or employers surveyed. A global employer survey and a US occupation outlook are not the same population. Next, sort the figure. A jobs-at-risk estimate is exposure. BLS outlooks are projections. A payroll or employment change is observed. AI taking over jobs statistics that skip this step mix all three. Check the year of any AI impact on jobs statistics too. A forecast made before recent tools shipped may be stale. For the wider debate, see the guide on [whether AI will replace jobs](https://nexumautomations.com/guides/will-ai-replace-jobs). ## What these figures do not tell you * Exposure Tasks that could change, not jobs lost. * Expectations What employers think will happen. * Young workers A low job-finding rate, not layoffs. Exposure is not loss. A task that could be automated may stay with a person for cost, trust or legal reasons. The Dallas Fed also notes a fall in employment for workers under 25 that reflects a low job-finding rate for young entrants, not layoffs. That is a different story from workers being replaced. Surveys of employers record expectations, and expectations can be wrong. Studies run by vendors lean toward their own product. Most of these numbers say nothing about wages, hours or job quality, only counts. Treat any single number as one reading, not the verdict. ## What it means for a 10-500 person company 1. List the tasks Note what each role spends hours on. 2. Measure Record time and cost today. 3. Pick one Start where hours are highest. 4. Track results Compare with the baseline. For a mid-size firm the practical question is tasks, not headcount. Half of small-business workers already use AI, but 64% of them use it for drafting and summarizing, and only 6% to run workflows with minimal human involvement. So most change today is faster individual work. A first step is listing which tasks in each role eat the most hours, then measuring them. The [AI adoption guide](https://nexumautomations.com/guides/ai-adoption) shows how to pick one. Roles also shift in what they do. MIT Sloan found firms with big AI increases grew employment faster, so planning for new tasks matters as much as planning for fewer. Methodology ## How we chose these numbers 1. Only figures from government data, universities, central banks and large research bodies are included. 2. Each figure keeps its source's meaning, with publisher, year and the group measured. 3. Exposure, projection and observed change are labeled apart and never merged. 4. Company announcements of job cuts are left out. 5. The page is updated when a newer release replaces a figure. FAQ ## AI job loss questions * ### What do AI replacing jobs statistics show so far? Observed change is small and uneven. In the sectors most exposed to AI, Dallas Fed data show employment down 1% since late 2022\. MIT Sloan finds the effect works at the task level more than at the whole-job level. * ### How many jobs will AI displace? No one knows. The World Economic Forum's survey projects 92 million displaced and 170 million created by 2030, a net gain of 78 million. It is a forecast from employers, so treat it as a range of expectations. * ### Are AI job loss statistics based on real layoffs? Mostly not. Many roundups of AI jobs statistics, AI taking jobs statistics or AI job replacement statistics mix forecasts, expert opinion and announced cuts. Measured sources such as the Dallas Fed use employment data and say entry-level weakness reflects slow hiring of young workers. * ### What do AI unemployment statistics say about young workers? The Dallas Fed links lower employment among those under 25 to a low job-finding rate for young entrants, not layoffs. The Fed also warns that displaced workers may face a stretch of higher unemployment while they search. * ### Which jobs are most at risk from AI? Exposure data and projections differ. The BLS projects little or no change for paralegals and legal assistants, with AI limiting demand, while projecting growth for lawyers and data scientists. Exposure is not the same as projected loss. * ### Does AI help or hurt wages? Mixed evidence. MIT Sloan found high-wage roles heavily exposed to AI grew their share of employment by about 3%, while top-paying roles inside firms fell 3.5%. Stanford's AI Index says AI mostly boosts productivity and narrows gaps between low- and high-skilled workers. * ### What do jobs lost to automation statistics show for small businesses? Automation is still rare there. In a 2026 survey of small-business workers, only 6% of AI users automate workflows with minimal human involvement, while 64% use AI for drafting, summarizing and brainstorming. --- Source: https://nexumautomations.com/statistics/ai-job-displacement Free AI diagnosis (30-minute call): https://nexumautomations.com/#book - Services: https://nexumautomations.com/services - Guides: https://nexumautomations.com/guides