Net Effects: Balancing AI-Related Job Losses with AI-Created Roles by Sector and Region

Net Effects: Balancing AI-Related Job Losses with AI-Created Roles by Sector and Region

Artificial intelligence (AI) is rapidly reshaping work. On one hand, many routine tasks – from data entry to customer support – can now be automated, leading firms to cut staff. On the other hand, new AI-intensive roles are emerging, such as data annotators, AI trainers, and machine-learning engineers. Analysts and surveys paint a mixed picture. For example, the World Economic Forum’s 2025 Future of Jobs report projected that by 2030 AI could create about 170 million new roles while displacing 92 million, yielding a net gain of ~78 million jobs globally (arstechnica.com). But most of those gains and losses are expected over many years. In the near term (through mid-2026), the effects are more modest and uneven by industry and region.

This article estimates AI’s net impact on jobs through June 2026 by combining reported layoffs with new AI-related job creation. We use talent supply-demand studies, cloud spending trends, and enterprise surveys to gauge where AI is fueling growth in jobs and where it’s cutting them. We break down results by sector and region, pointing out fields where new AI roles largely offset cuts, and others where losses dominate. Finally, we discuss how displaced workers fare – how long it takes them to retrain and reach full productivity in new jobs – and suggest actions for workers and employers.

AI-Related Job Losses: What We Know

Across industries worldwide, companies have announced significant layoffs in recent years, often citing AI or automation as a factor. An independent jobs tracker tallied over 1.4 million jobs reported displaced by AI-related restructuring as of early 2026 (aijobimpact.org). For example, many large tech firms cut thousands of roles: tech companies (including Amazon, Google, Microsoft) announced roughly 100,000 jobs losses since 2023 (www.replaced-jobs.com). Telecommunications and manufacturing companies (e.g. BT Group, Nokia, Foxconn) also reported massive cuts. In Q1 2026 alone, tech industry layoffs reached about 78,600 globally, nearly half of which firms directly attributed to AI and automation (www.tomshardware.com).

In finance, AI is similarly disrupting jobs. Banking and insurance firms worldwide are automating analysis and back-office work. One sector analysis warned that millions of U.S. finance jobs could be at risk by 2027 (e.g. ~890,000 in banking, 1.4 million in insurance, 1.08 million in accounting) (www.aiexposure.org). Morgan Stanley data finds that some AI-investing firms are already using “AI as a license to cut headcount” – for example, UK firms in key industries saw net job declines (23% of roles eliminated vs 15% new hires) while U.S. firms surveyed actually saw net growth (17% lost vs 19% hired) (www.itpro.com).

However, not all projections show massive declines. Goldman Sachs research suggests only modest overall impacts. They estimate U.S. unemployment might rise by only ~0.5 percentage point during the AI transition, and at most 2–7% of U.S. jobs would be displaced if AI efficiency gains led to labor cuts (www.goldmansachs.com). Similarly, analysts at Gartner predict no “jobs apocalypse,” but foresee widespread job reconfiguration. Gartner estimates that from 2028 onward some 32 million jobs per year will be reshaped or redesigned due to AI – meaning roles will evolve, not entirely vanish (www.itpro.com).

In short, multiple sources confirm significant, but uneven job losses. Tech and telecom sectors have seen large cuts already (www.tomshardware.com). Industries like manufacturing and finance are also automating aggressively. By mid-2026, trackers suggest North America and Asia have seen net job losses on the order of hundreds of thousands (e.g. ~385k U.S. roles displaced vs 238k new hires, net –147k) (aijobimpact.org). Europe shows similar net losses (~315k displaced vs 212k created, net –103k) (aijobimpact.org). Smaller markets like South America show lighter net losses, and Africa stands out with a net gain (84k new vs 65k displacements, +19k) (aijobimpact.org). (These figures come from comprehensive trackers of published layoffs and hires.)

In practical terms, typical AI-driven job cuts affect roles heavy on routine or data tasks: customer service agents, entry-level analysts, back-office clerks, etc. McKinsey projects that by 2030, tens of millions of such customer-facing, clerical, and production jobs in the U.S. could decline, even as high-skill jobs persist or grow (www.mckinsey.com). Early 2025 survey data supports this: IT roles like radiologists or recruiters (where AI augments human experts) held up or even grew, but roles like IT support or secretarial work (where AI can automate routine tasks) saw slower growth (www.itpro.com).

AI-Related Job Gains: New Roles on the Rise

Alongside losses, AI is creating many new positions – not just better versions of old jobs, but entirely new kinds of work. Think of data labelers for machine learning, AI software engineers, prompt engineers, AI ethics specialists, and so on. Surveys and talent trackers confirm surging demand for AI skills. For example, a global hiring report found demand for “AI training” roles (like data annotators) grew 283% in 2025 alone, and there are now over 70,000 workers worldwide doing AI training tasks (www.itpro.com). AI job postings (for machine learning engineers, data scientists, etc.) have been growing roughly 8 times faster than the overall job market (www.itpro.com), and the number of posted “AI jobs” nearly doubled from 2024 to 2026 (www.itpro.com).

These new roles cluster in certain sectors. According to PwC’s recent analysis, technology, media/telecom, and professional services have seen the largest upticks in AI job creation (www.itpro.com). Think cloud providers, software companies, advertising/media firms, and consultancies – areas deeply investing in AI projects. In contrast, sectors like education or general healthcare have shown relatively little growth in AI job postings so far (www.itpro.com). (PwC also found that “reskilled” entry-level jobs – where junior roles now demand leadership or creative skills – grew 35% since 2019, while other entry-level roles shrank 10% (www.itpro.com).)

On the worker supply side, talent shortages signal rapid role creation. In the UK, demand for AI professionals is exploding: by 2028 the UK may need ~300,000 AI-skilled workers but produce only ~137,000 domestically (www.itpro.com). This gap is leading firms to hire globally. One analysis notes the UK already accounts for 20% of cross-border tech hires (double the U.S.) as companies fill AI roles with international talent (www.itpro.com). Globally, platforms for remote work report that AI-related roles are among the fastest-growing. (For instance, a large jobs site found that from 2025 onward, average salaries for AI-related postings climb well above typical rates (www.itpro.com).)

Another proxy for AI’s ramp-up is corporate spending. Global cloud infrastructure spending – a rough stand-in for enterprise AI use – skyrocketed in 2025. Cloud revenues grew ~24% to nearly $400 billion, and are on track to exceed $500 billion in 2026 as companies deploy AI in production (www.techradar.com) (www.techradar.com). That spending typically brings more IT jobs: cloud architects, data engineers, Nvidia-GPU specialists, etc. Indeed, companies investing in AI often report hiring more, not fewer, people. For example, an EU study cited by Tom’s Hardware found that companies deploying AI tended to expand headcount (www.tomshardware.com). IBM’s CEO recently noted that his company tripled entry-level hiring in early 2026 despite AI advances, because human skills remain crucial (www.tomshardware.com).

In quantitative terms, one global tracker combining announced hires suggests roughly 1.02 million new AI-related jobs have been created (mostly in IT and tech) versus 1.48 million displacements, for a net –457 thousand worldwide through mid-2026 (aijobimpact.org). By region, this tracker reports Asia with ~605k displacements vs 410k new (net –195k), North America ~385k vs 238k (–147k), Europe 315k vs 212k (–103k), Latin America 92k vs 74k (–18k), and Africa 65k vs 84k (+19k) (aijobimpact.org) (aijobimpact.org). The African gain is notable – it may reflect smaller-scale automations plus growing demand for IT roles (for example, outsourcing of data labeling or software development). In short, AI job growth has been concentrated in tech/knowledge fields and certain regions.

Sectoral Balances: Who Gains vs. Who Loses

Putting the pieces together, we see winners and losers by industry. Broadly:

• Technology & Software. Tech firms face heavy AI-driven restructuring, but they also create the most new AI roles. Listings for AI engineers, cloud architects, and ML experts are booming. In the US, regions with many tech workers (Silicon Valley, Seattle, Austin) saw ~156% job growth in AI skills (www.itpro.com). Overall, many tech companies report shortages of AI talent, meaning they are hiring. Yet tech layoffs have also been large (78k in Q1 2026 (www.tomshardware.com)). Net impact: mixed – some subfields (AI research, cloud services) are expanding jobs, while others (legacy enterprise software support) are contracting.

• Finance and Insurance. Automation is aggressively targeting back-office and analyst tasks. Many banks are using AI for contract analysis, risk modeling, and customer service. Surveys predict large “jobs at risk” in banking, accounting, loan processing and insurance underwriting (www.aiexposure.org). Indeed, several financial institutions reported cuts among analysts or clerks. On the flip side, banks also hire in AI compliance, data crypto, and fintech roles – but not enough yet to offset large cuts. Net effect so far appears negative in financial sectors. (For example, UK data show finance-heavy industries had more cuts than AI-driven hires (www.itpro.com).) Goldman Sachs estimates only 2.5–7% of U.S. jobs would be cut even under broad AI use, but these are heavily concentrated in financial admin and asset management (www.goldmansachs.com).

• Manufacturing & Logistics. Assembly and warehousing jobs are increasingly automated. Companies like Foxconn and 3M announced workforce reductions in 2024-25 that they partly attributed to efficiency drives (including AI robotics) (www.replaced-jobs.com) (www.replaced-jobs.com). However, these industries also need new skills to maintain and program robots and AI-enabled machines. For example, carmakers use AI for design and quality control, requiring roles in AI maintenance. In aggregate, many manufacturing sub-sectors show net slight losses in the short run, though this varies by locale. Early data cited by Morgan Stanley found the automotive sector among the hardest hit globally (www.axios.com), but a detailed US impact involves both automation and broad tech shifts.

• Healthcare. AI in healthcare (like image analysis or documentation) is growing, but so far fewer layoffs are attributed directly to AI. Hospitals and clinics still need large human workforces (nurses, care providers); AI tools mainly augment them. As one report noted, healthcare has the lowest share of new AI jobs among major sectors (www.itpro.com). Thus, healthcare manufacturing (big equipment) might see some losses, but patient-facing healthcare likely remains stable or grows. The net is likely neutral to slightly positive for healthcare overall.

• Retail, Hospitality, and Services. E-commerce and automation have been changing roles for years. In retail (especially supply chains), robotics and AI are cutting some roles (like warehouse sorting) but spurring others (like AI-driven inventory management). Service jobs (hotels, restaurants) see early automation via kiosks and chatbots. Overall, the pace of AI job cuts here is gradual. Interestingly, some of the roles being lost (e.g. entry-level clerks) are being partly offset by demand for higher-skilled workers (data analysts, digital marketers). In sum, retail and services show mixed results – with a gradual shift in skill needs, but not a dramatic net loss so far.

In short: tech/media/communications and professional services (consulting, enterprise IT) are adding AI-related jobs rapidly (www.itpro.com), often roughly balancing any reductions. In contrast, finance, routine admin, and manufacturing have seen more cuts and fewer new positions yet. Sector productivity reports bear this out: companies heavily exposed to AI report strong productivity growth and are still hiring (for high-skill roles) (www.itpro.com), whereas sectors just beginning to automate are cutting deeper.

Regional Balances: Where Losses Outweigh Gains (and Vice Versa)

The net impact also varies by region. High-tech, high-service economies like the U.S., UK, and Western Europe are on the leading edge of AI adoption – and thus early job disruption. The US shows substantial displacement in tech and office sectors, but also has many new AI roles in tech hubs. Interestingly, surveys indicate U.S. companies as a whole saw a net increase in jobs from AI adoption: one study found 19% U.S. firms reported new hires versus 17% job losses (www.itpro.com). This reflects heavy growth in tech and a flexible labor market. Goldman Sachs similarly finds a moderate U.S. employment risk (up to 2.5–7% in worst cases) (www.goldmansachs.com).

In Europe, the picture is mixed. Many large manufacturers and banks are automating (e.g. major EU banks have announced layoffs). Surveys in early 2026 showed Europeans expecting strong tech hiring: one report found European employers plan 27% growth in tech roles in 2026 and 17% in 2027 due to AI (www.techradar.com). However, European job markets (which were slower to adopt AI than Silicon Valley) probably face more upfront losses in routine jobs, with gains coming slightly later. The Linux Foundation report notes that overall Europe may see net job creation in tech, but half of firms worry AI will disrupt work (www.techradar.com).

By contrast, emerging economies and developing regions are seeing relatively smaller cutbacks. For instance, Africa (with fewer routine white-collar jobs to automate) is showing net AI-job gains in trackers (aijobimpact.org). Some governments in Asia (e.g. India, China) are heavily investing in AI skill programs. Notably, one tracker shows India with a net gain of ~40k AI jobs as of 2026 (aijobimpact.org). China’s economy is more mixed; media stories report millions of routine jobs at risk and heavy emphasis on AI education, so eventual losses could be large, but the technology sector is also booming.

Key takeaway on regions: Western tech-centric economies (North America, Europe) are seeing both high AI adoption and significant displacement – roughly balancing new tech jobs with automations so far. Some developing regions (Africa, parts of Asia) are on the early side of adoption, so a surge in new tech roles (outsourced AI tasks, infrastructure) temporarily outpaces losses. In real numbers from trackers: Europe and North America each show net losses on the order of 100–150k jobs by mid-2026 (aijobimpact.org), Asia shows a larger net loss (~195k) due to scale, while Latin America’s net loss (~18k) is smaller and Africa has a net gain (aijobimpact.org) (aijobimpact.org).

Time-to-Productivity for Reemployed Workers

An important part of the net impact is how quickly displaced workers can find new jobs and regain productivity. Even if overall jobs balance out, it takes time and training for people to transition. Research shows this transition is often slow. For example, U.S. retraining programs like WIOA (workforce innovation grants) rarely shifted participants into less AI-exposed occupations – many returned to similar fields (deepmind.google). When workers from high-automation-risk jobs do retrain, their earnings have risen: one study found displaced high-exposure workers earned only ~$900 per quarter before 2020 but ~$2,900 by 2022–24 after re-training, largely by moving into lower-risk roles (www.nber.org). This suggests significant wage (and presumably productivity) recovery, but it typically comes after intensive training and job search, which can take months or years.

Programs like apprenticeships tend to have the quickest payoff. The DeepMind analysis noted that employer-led apprenticeships have “the highest incidence of success” in moving people into good jobs (deepmind.google). In general, experts emphasize upskilling: Gartner projects that 150,000 workers per day will need retraining in coming years to keep up with AI-driven role changes (www.itpro.com). Studies also warn that short-lived cuts followed by insufficient training can be counterproductive – losing entry-level workers too soon destroys on-the-job learning pipelines (www.tomshardware.com).

For individual workers, the message is that acquiring AI-relevant skills is critical. Workers with AI or “human-intensive” skills (leadership, creativity, empathy) see a wage premium (www.itpro.com). Employers are increasingly hiring on skills over degrees, especially for AI and tech jobs (goatstack.ai). But this retraining takes time – workers often need many months of education or bootcamps to be fully productive in a new AI role. Even then, performance may lag initially while one learns company specifics.

In sum, the lag from a lost job to full productivity in a new role is substantial. On the order of months of training plus an adjustment period on the job. That means even if overall employment eventually balances, transitional unemployment can rise in the short term. In practice, studies find that displaced workers often face initial wage dips and an extended job search. Policy reports caution that without proactive help, aggregate productivity gains from AI will materialize only slowly, as people resettle into new roles (deepmind.google) (www.nber.org).

Conclusion and Actionable Advice

Bottom line: By mid-2026, AI is causing both job cuts and job creation in most sectors. The net effect is a moderate job loss in many advanced economies, with some offsetting gains in AI-intensive fields. In high-tech, finance, and manufacturing, the tide of automation has outpaced new AI hiring so far. But in technology, consulting, and specialized niches (like AI engineering, cloud services), job growth is robust. Some regions (notably parts of Asia and Western economies) are net-negative, while others (like Africa) are net-positive. Critically, the new roles often require different skills, so displaced workers must invest in training.

For workers and employers, the key advice is to plan and progress early:

• Invest in skills now. Workers should build AI literacy and complementary skills (data literacy, coding basics, or very importantly, human skills like problem-solving). Training programs – whether bootcamps, online courses, or apprenticeships – are crucial. Companies and governments should ramp up reskilling programs. Internal training (upskilling on the job) is often faster and cheaper than recruiting new talent (www.techradar.com), so firms should teach current employees how to work with AI.

• Draw on global talent. Regions facing AI talent shortages (e.g. the UK) can hire internationally or partner remotely to fill roles, as some are doing (www.itpro.com). Policies that ease cross-border hiring for tech skills will help alleviate stiff competition.

• Track where new roles emerge. Industries and regions looking to offset losses should incentivize companies to create those roles (for instance, governments could support startups in AI, establish tech hubs, or promote cloud computing industries). We’ve seen that cloud and AI spending growth (e.g. $500B+ global cloud spend in 2026 (www.techradar.com)) brings jobs; policies that encourage this investment can spur job creation.

• Support transitions. Public planning should anticipate “reconfigurations” famous in AI forecasts (www.itpro.com). This means beefing up unemployment support, job counseling, and tailored training. Studies find people often end up in similar fields, so career services should help match skills to evolving roles, not just push generic upskilling (deepmind.google).

• Focus on human-plus-AI roles. Finally, remember that many surviving and new jobs will combine AI with human skills. For example, radiologists now use AI tools but still need a human for diagnosis decisions, and customer service reps increasingly handle only the hardest cases. Emphasizing uniquely human skills (leadership, creativity, caregiving, ethical oversight) is a way to make sure your role stays in demand (www.itpro.com) (www.itpro.com).

The rise of AI won’t eliminate the need for hard work in finding and learning new jobs. But by preparing ahead – continuously learning and focusing on areas where human judgment complements AI – workers and economies can mitigate short-term losses. Likewise, businesses that blend AI with human strengths stand to become more productive while maintaining strong headcounts (www.itpro.com) (www.techradar.com). Overall, the evidence suggests AI will reshape the labor market, but not wipe it out. Planning for a gradual transition – with a focus on retraining and new workforce strategies – is the most realistic path.