The Scary Reality of AI: How It’s Changing the Workforce and the World

AI is often hailed as a breakthrough technology, offering countless benefits across industries, from healthcare to transportation. But while we can marvel at the possibilities, there’s a darker side to the rise of AI that we can’t ignore. The rapid development of Agentic AI—the kind of AI that can make independent decisions and take action without human oversight—presents an alarming reality: millions of jobs could be displaced in the coming years.

Think about it: automation has already started to creep into industries like manufacturing, customer service, and even transportation. Driverless cars, AI-powered chatbots, and robotic assembly lines are just the beginning. In the next 5 to 15 years, these technologies could lead to mass unemployment as robots and AI systems replace human workers. The most vulnerable jobs? Those that rely on repetitive tasks, like call center employees, cashiers, and delivery drivers. But this isn’t just a problem for low-wage workers—it’s a challenge for everyone.

As we look to the future, we face a rapidly changing job market, where the very skills we’ve spent years developing may become obsolete. The scary part? Not everyone will be prepared for this shift. While new job opportunities will certainly emerge in fields like AI development, healthcare, and cybersecurity, they’ll require a completely different set of skills. And for those unable to make the transition, the future looks uncertain.

The question we must ask ourselves is this: Will AI be the job creator of tomorrow, or will it leave millions struggling to find their place in a world where machines are increasingly in charge? The reality is unsettling, and if we don’t act now—by investing in education, retraining programs, and social support systems—we risk facing a future where the very technology meant to help us may instead leave us behind.

It’s time we start thinking seriously about the impact AI will have on our lives, our livelihoods, and the future of work itself. The scary reality of AI is not just that it’s changing the way we work—it’s changing the way we live. Let’s make sure we’re ready for what comes next.

AI and Employment Projections: USA, UK, and Global Outlook

Current Labor Force and Unemployment

United States: The U.S. labor force includes roughly 168 million people, with unemployment around 4% in early 2025​

bls.gov. This corresponds to about 6.8–6.9 million Americans unemployed out of ~162 million employed​

bls.gov. The job market has been tight recently, with unemployment near historic lows (3.8%–4.1% over 2023–2024)​

bls.gov. However, these low unemployment figures form the baseline against which AI-driven changes will be measured.

United Kingdom: The UK has about 35–36 million people in its workforce. As of late 2024, 1.56 million Britons were unemployed, a rate of 4.4%, up from 3.9% a year earlier​

commonslibrary.parliament.uk. UK unemployment remains low by historical standards, but slight rises in 2024 hint at a cooling labor market​

commonslibrary.parliament.uk. This context of relatively low joblessness could shift as automation advances.

Worldwide: The global labor force is approximately 3.5–3.6 billion people. The worldwide unemployment rate in 2024 is estimated around 5.8%, with about 211 million people unemployed​

ilostat.ilo.org. Emerging economies tend to have higher underemployment, while many advanced economies face labor shortages. This global backdrop means over 200 million jobseekers even before any large-scale AI displacement, highlighting the importance of how AI will affect employment internationally.

Expected Job Displacement from AI Automation

United States (5, 10, 15-Year Outlook): Multiple studies estimate substantial U.S. job disruption from AI and automation over the next decade. By 2030 (5 years), around a quarter of U.S. jobs (36 million positions) could face high automation risk​

davidrousefaicp.com. Sectors like office support, manufacturing production, transportation, and food service are especially vulnerable to AI-driven task replacement​

davidrousefaicp.com. For example, self-driving vehicle tech could automate trucking and delivery jobs, and AI chatbots may handle routine customer service calls. By 2035 (10 years), up to 38–40% of U.S. jobs might be impacted in some way by AI according to some forecasts​

futurism.com. However, “impacted” includes both automation of tasks and transformation of roles – it doesn’t mean all those jobs vanish. Looking out to 2040 (15 years), forecasts diverge. Some tech visionaries like Kai-Fu Lee have warned that 40% of jobs worldwide could be done by machines by the mid-2030s​

futurism.com, implying a significant impact on the U.S. as well. More conservative analyses suggest the U.S. will experience continued job churn rather than a sudden employment collapse – with many workers transitioning occupations rather than becoming permanently unemployed​

mckinsey.com​

mckinsey.com. In a worst-case scenario, if AI advances extremely fast and primarily replaces (rather than complements) human labor, U.S. unemployment could rise notably by the late 2030s. But mainstream economic projections (including official U.S. Bureau of Labor Statistics outlooks) still expect net job growth through the next decade​

bls.gov, indicating that any AI-driven layoffs may be offset by job creation elsewhere under current trends.

United Kingdom (5, 10, 15-Year Outlook): Automation is also poised to reshape the UK job market, though moderate in the near term. By 2030, around 30% of existing UK jobs could be impacted by AI/automation according to PwC analysis​

pwc.com. That equates to millions of roles affected, especially in sectors like clerical work, retail, and routine service jobs. However, near-term unemployment changes may be modest. One UK study’s “best guess” scenario (from the Tony Blair Institute) estimates AI will raise UK unemployment by only about 30,000 by 2030​

hkifoa.com – a relatively small bump given ~1.5 million Brits are already unemployed. By 2035, the share of jobs affected grows; up to one-third of UK jobs could see significant task automation by the mid-2030s​

pwc.com. Many workers will need retraining as administrative and data-processing roles decline. Surveys suggest British employers anticipate gradual change: in one poll, 20% of employers expected under 10% of their workforce to be replaced by 2034, and another 19% foresaw no AI-related staff reductions​

britsafe.org. By 2040, impacts depend on how technology and the economy evolve. In an adverse scenario, AI could potentially double the UK’s unemployment (from ~1.4 million to almost 3 million jobless by 2040) if new jobs don’t materialize​

hkifoa.com. But analysts consider this doomsday outcome unlikely​

hkifoa.com. A more probable scenario is a peak increase of a few hundred thousand additional unemployed (the same study’s baseline projected a peak of +340,000 UK unemployed around 2040) before the labor market adjusts​

hkifoa.com. In short, the UK may see noticeable but manageable displacement over 10–15 years, concentrated in particular job types, rather than an across-the-board employment collapse.

Worldwide (5, 10, 15-Year Outlook): Globally, the scale of AI-driven displacement could be massive, but spread over time. By 2030, studies estimate anywhere from ~85–800 million jobs worldwide could be displaced or transformed by automation and AI. A widely cited McKinsey Global Institute report estimated 400–800 million individuals may need to change occupations globally by 2030 due to automation​

mckinsey.com. The World Economic Forum (WEF), surveying employers in numerous countries, projects about 92 million jobs will be displaced by 2030 due to AI, robotics, and other shifts – but concurrently 170 million new jobs will be created​

businessbecause.com. That implies a net gain of +78 million jobs globally by 2030 (roughly 7% of the workforce) if their forecast holds​

businessbecause.com. In other words, even though tens of millions of workers worldwide may be replaced or redeployed by AI in high-risk sectors (manufacturing, clerical, call centers, etc.), even more jobs might emerge in sectors like tech, green energy, and care economy. By 2035 (10 years), the reach of AI is expected to broaden. The IMF reported that about 40% of jobs globally could be affected in some way by AI (positive or negative) by the mid-2030s, with higher exposure (up to 60% of jobs) in advanced economies​

politico.eu. Many developing countries will face automation in industrial and service roles, but also could gain jobs if AI-driven growth raises global demand. By 2040 (15 years), long-term global predictions are speculative – some futurists argue we could approach an era where nearly half of all work tasks worldwide are automated around that time frame. If AI and robotics progress unabated, large developing economy workforces (e.g. in China, India) might see manufacturing and back-office jobs shrink substantially by 2040. On the other hand, entirely new industries could exist by 2040 (in areas like advanced robotics maintenance, AI-driven healthcare, space tech, etc.) employing millions. The net global unemployment effect will hinge on how quickly displaced workers can transition and how much new economic activity AI generates. History suggests that labor markets ultimately adapt to major technological shifts – but the transition can be painful if it occurs faster than education/training systems and policy safety nets can handle​

mckinsey.com​

mckinsey.com.

New Job Creation from AI and Emerging Industries

Crucially, AI is not only a job destroyer – it’s also a job creator. Many forecasts highlight that alongside displacement, new occupations and industries will emerge:

• Technology Sector Jobs: The AI industry itself is booming. Demand for AI developers, machine learning engineers, data scientists, and cybersecurity experts is growing rapidly. In the U.S., tech employment is projected to keep rising; for example, the BLS projects software development and data-related roles will see above-average growth this decade (helping offset losses in routine office jobs). Globally, roles like “AI and machine learning specialist” and “data analyst” have been among the fastest-growing job titles in recent years​wallyboston.com. These roles barely existed 15 years ago, underscoring how technology creates entirely new careers. In fact, McKinsey notes that by 2030 roughly 8–9% of labor demand could be in new job categories that don’t exist today – similar to how the internet gave rise to web developers and digital marketers​mckinsey.com.
• Green Economy and Infrastructure: AI’s spread coincides with other transformations like the shift to clean energy. Governments investing in renewable energy, electric vehicles, and infrastructure are creating new jobs (e.g. solar panel installers, EV maintenance, smart grid engineers). These trends are expected to continue through 2030. In the U.S., for instance, federal climate and infrastructure investments will likely produce a modest net job gain – employing more people in construction, EV production, and public transit even as fossil-fuel-related jobs decline​mckinsey.com. Many of these new roles (battery technicians, wind turbine operators) benefit from AI tools but are fundamentally new workforce needs.
• Healthcare and Education: As populations age and AI enables improvements in medicine, healthcare jobs are set to expand worldwide. AI can assist doctors and nurses (for example, by automating routine diagnostics), but it also increases demand for tech-skilled healthcare workers (to implement AI systems) and caretakers (as healthcare becomes more accessible). McKinsey expects increased demand for healthcare workers in the coming years despite automation​mckinsey.com. Education could similarly see new roles – AI tutors, education technologists, etc., as society invests in re-skilling workers for an AI-driven economy.
• Roles Complementing AI: Rather than replacing humans entirely, AI often works alongside humans, creating complementary jobs. For example, the rise of AI in business has led to roles like AI trainers (people who label data or teach AI systems), prompt engineers (specialists who craft inputs for generative AI), and AI ethics officers to oversee algorithmic fairness. These niche jobs barely existed a few years ago and are now increasingly common in organizations deploying AI. Similarly, robotics in factories create demand for robotics technicians and maintenance specialists. Cybersecurity is another growth area – as companies digitize and use AI, securing those systems becomes vital, fueling jobs in cyber defense. All these are part of the “innovation ripple effect” of AI, where each new technology triggers ancillary employment needs.

Accordingly, the net impact of AI on jobs is a balance between the losses in certain tasks and the gains in new functions. The World Economic Forum’s global employer survey found that while millions of jobs will be made redundant, even more could be created in fields like big data, AI, engineering, digital marketing, and so on​

businessbecause.com. Their analysis for 2020–2030 anticipates 170 million new jobs globally, versus 92 million eliminated, with net positive job growth concentrated in sectors that adopt technology​

businessbecause.com. To realize these gains, however, the workforce must have the skills to fill the new roles – which is why there is heavy emphasis on training and education in tandem with AI adoption.

Net Impact on Employment Levels

Taking both automation and new job creation into account, experts have a range of predictions for net employment effects in the next 5, 10, and 15 years:

• Short-term (Next 5 Years): The consensus is that by 2025–2030, AI will cause significant labor market churn but not necessarily a net spike in unemployment. Employers anticipate around 23% of current jobs will change (either grow or decline) by 2027 globally​wallyboston.com. Many jobs will be redefined rather than completely lost. In advanced economies like the US and UK, unemployment rates might remain relatively stable if new hiring in tech, healthcare, and other growing areas absorbs workers leaving shrinking occupations. For instance, the WEF expects a roughly balanced turnover in the workforce through 2030, even projecting a net addition of jobs worldwide​businessbecause.com. That said, certain regions and demographics may feel pain sooner – e.g. manufacturing-heavy towns or older workers with skills that don’t match emerging jobs could see higher localized unemployment. Overall, through 2030, AI is more likely to reshape jobs than to cause sustained mass joblessness, provided the economy continues to grow and innovate.
• Medium-term (Next 10 Years): By 2035, the cumulative effects of AI automation will be larger. It’s during this period that some economists warn net impacts could turn negative if job creation lags. One projection for the UK “whirlwind” scenario showed unemployment possibly doubling by 2040 in a case where AI adoption is rapid and largely labor-substituting​hkifoa.com. Similarly, if 40% of world jobs are truly automated by ~2035 as pessimistic forecasts suggest, many countries could experience higher jobless rates. However, a more mainstream view is that productivity gains from AI will generate new demand that offsets these losses. History shows technology-driven productivity boosts often lead to higher GDP and new industries, which eventually employ those displaced. For example, the adoption of AI across industries is expected to raise global GDP by about 14% by 2030 (an increase of $15 trillion) according to PwC – this growth creates purchasing power and business opportunities that spawn jobs​businessbecause.com. OECD analyses similarly conclude that while ~14% of jobs could be fully automated, most workers will transition to new roles and total employment in 2030s may be similar or higher than today, albeit with a different mix of jobs. In practice, we might see unemployment tick up in the mid-2030s in some countries during the adjustment period, but not skyrocket, especially if new job creation continues at pace.
• Long-term (Next 15+ Years): By 2040 and beyond, uncertainty is greatest. If AI reaches an advanced stage (e.g. truly autonomous systems performing a majority of current human work), there could be a more dramatic decoupling of employment from economic output. Some technologists envision a future by 2040 or 2050 where human labor demand falls significantly in traditional sectors – raising questions about sustaining low unemployment. On the other hand, optimists argue that by the 2040s, entirely new categories of work will employ people (just as nobody in 1980 could predict the web or app economy jobs of today). The Tony Blair Institute’s analysis suggests even with robust AI adoption, the UK’s unemployment peak effect might be on the order of only a few hundred thousand extra unemployed, and that effect “unwinds over time” as the economy adjusts​britsafe.org. In other words, after an initial disruption, employment levels could normalize at a new equilibrium. Globally, the net effect by 2040 will hinge on policy choices and how broadly the gains of AI are shared. With “sufficient economic growth, innovation, and investment, there can be enough new job creation to offset the impact of automation” one major study concludes – but without those, some economies could face job shortages​mckinsey.com​mckinsey.com. In summary, the net impact of AI on 15-year horizons ranges from mildly positive (if managed well, yielding higher productivity and new industries) to mildly negative (if adoption outpaces worker transition). A collapse in global employment is not widely expected, but neither is it guaranteed that every displaced worker finds a new job immediately – there will be winners, losers, and a need for proactive adaptation.

Industry Hotspots: Jobs Most at Risk vs. Jobs on the Rise

AI will not affect all industries equally – some roles face high risk of automation, while others are likely to see increased demand. Across the USA, UK, and globally, several high-risk industries stand out:

• Manufacturing & Warehousing: Roles involving routine manual tasks or predictable workflows are prime targets for robotics and AI. Factory assembly line workers and machine operators could be partially replaced by industrial robots and AI-driven quality control systems. A 2019 analysis warned that up to 20 million manufacturing jobs worldwide could be lost to robots by 2030, many in developing countries’ factories. Warehouse and fulfillment center jobs (like packers, forklift operators) are increasingly automated by AI-guided robots and inventory systems. These sectors have already seen automation for decades, and AI will accelerate the trend in the next 5–15 years.
• Clerical, Administrative & Customer Service: Jobs that are fundamentally about processing information or routine interactions are very exposed to AI. This includes secretaries, data entry clerks, bank tellers, bookkeeping clerks, and call-center agents. In the UK, ONS data identified waiters, retail cashiers, and clerical assistants among those with the highest automation probabilities​theguardian.com. AI chatbots and virtual assistants can handle basic customer inquiries; software bots can perform data entry and bookkeeping faster than humans. As a result, clerical and administrative jobs are projected to decline quickly in the coming years​businessbecause.com. For example, many banks are shifting to AI-powered online services, reducing the need for branch staff. These roles employ millions (especially women, who disproportionately hold office support jobs), so their contraction is a significant source of potential unemployment due to AI.
• Transportation & Logistics: Self-driving technology and algorithms threaten jobs like truck drivers, taxi drivers, and delivery couriers. By the mid-2030s, autonomous vehicles and drones could make a serious dent in these occupations. Some estimates suggest long-haul truck driver positions may peak and start declining as early as the 2030s once self-driving trucks become commercially viable. Likewise, warehouses using self-driving forklifts and delivery bots will need fewer human movers. Though full autonomy at scale is taking longer than expected (safety and regulatory hurdles persist), incremental AI improvements (driver-assist, routing optimizations) are already reducing labor needs. The Brookings Institution noted transportation and material moving jobs have a high share of automatable tasks, putting them in the high-risk category by 2030​davidrousefaicp.com.
• Predictable Service and Sales Roles: Some service jobs with repetitive routines – for instance, fast-food workers (with AI-driven kiosks and kitchen robots), telemarketers, and insurance underwriters – face AI substitution. Retail sales could also be hit: AI-powered e-commerce and cashierless store technology (like self-checkout or Amazon’s AI stores) means fewer retail clerks. However, these changes may be gradual; customer preference for human service and the complexity of some tasks means not all such jobs will disappear by 2030, but their growth will slow.

On the other hand, certain job categories are poised for growth precisely because of AI and related technological advancements:

• Technology Development & Maintenance: As mentioned, AI specialists, software engineers, and IT professionals are in higher demand. For every algorithm deployed, there is need for people to design, train, and maintain it. The U.S. for example has a shortage of tens of thousands of AI professionals, which the government and industry are trying to address through education initiatives​hrmagazine.co.uk. Similarly, jobs in robotics maintenance, IT support for AI systems, and cloud computing are burgeoning. These roles tend to be high-skill and well-paid, absorbing some workers from shrinking fields (e.g. a factory worker retrained in robot maintenance).
• Healthcare & Caregiving: AI’s ability to augment healthcare (from diagnostics to administrative streamlining) will likely increase the efficiency of medical services, spurring demand to serve aging populations. We expect more nurses, medical technicians, and home health aides will be needed, not fewer – AI will handle some tasks, but rising healthcare needs mean humans will focus on advanced care and interpersonal aspects. An AI-assisted doctor can see more patients, but you still need more doctors globally to handle aging demographics. The WEF projects large job growth in healthcare and social assistance by 2030 as one of the key areas of new employment​mckinsey.com.
• Education & Training: Far from eliminating the need for teachers, AI could boost education roles. There will be demand for educators who can teach new digital skills, for AI literacy trainers, and for people to develop AI-driven educational content. Additionally, as workers face career shifts, the adult training and vocational education industry will grow (e.g. coding bootcamp instructors, corporate trainers specializing in AI tools). This sector might expand specifically to help re-skill those displaced – effectively creating jobs for those who train others in new skills.
• Creative and Complex Professions: Jobs that require creativity, complex problem-solving, and human judgment are relatively insulated and can even flourish with AI. Architects, scientists, engineers, strategists, and artists may use AI as a tool to enhance productivity, not as a replacement. For instance, an architect with generative design AI can take on more projects, possibly hiring additional assistants as their output expands. New creative tech fields (video game design, VR experience creation, digital marketing strategists) are likely to employ more people as digital content demand explodes. Notably, WEF surveys found “analytical thinking” and “creative thinking” are still the top skills needed in the workforce despite AI’s rise​wallyboston.com – indicating roles that leverage those human strengths will remain in demand.
• Public Sector and AI Governance: As AI becomes pervasive, governments and organizations will create roles for AI policy, regulation, and ethics oversight. We may see more jobs like AI auditors, compliance officers, and ethicists ensuring AI systems are fair and legal. This is an emerging niche but one likely to grow by the 2030s, contributing to job creation especially for those with interdisciplinary skills in law, sociology, and technology.

In summary, routine and repetitive jobs are at high risk, whereas jobs requiring human creativity, empathy, or complex decision-making are more secure – and many of those will grow, not shrink. High-risk industries will need to transition their workforce into the growth industries through retraining. If managed well, the expansion of “jobs on the rise” can compensate for losses in “jobs at risk,” keeping overall employment healthy.

Policy Responses and Mitigation Strategies

Governments and institutions are not idle in the face of potential AI-driven unemployment. A variety of policy responses are being discussed or implemented in the US, UK, and globally to mitigate the disruption:

• Workforce Retraining and Upskilling Programs: This is the most immediate and widespread response. Both the U.S. and UK governments are investing in programs to teach workers new skills for emerging jobs. The UK’s AI Opportunities Action Plan (2024), for example, emphasizes training “tens of thousands of AI professionals by 2030” to fill skill shortages​hrmagazine.co.uk. It calls for integrating AI into all levels of education and expanding digital apprenticeships to help workers transition into tech roles​hrmagazine.co.uk​hrmagazine.co.uk. In the U.S., policy think-tanks have proposed an “AI Adjustment Assistance” program modeled on Trade Adjustment Assistance, which provided training and income support to workers displaced by globalization​urban.org​urban.org. This would retrain workers who lose jobs to automation, helping them shift into in-demand occupations. Many companies are also launching their own upskilling initiatives; per the WEF, 77% of surveyed firms plan to retrain or reskill workers to help them work alongside AI by 2030​arstechnica.com. The clear policy trend is funding continuous learning so that the workforce can keep pace with technological change.
• Education System Reforms: Beyond short-term training, governments are looking at education curricula to prepare the next generation for an AI-infused economy. STEM (science, technology, engineering, math) education is being bolstered, and there’s growing focus on teaching digital literacy, coding, and AI basics from early ages. For instance, the UK is encouraging more AI-related university degrees and PhDs (after noting only ~46,000 AI-relevant graduates in 2022)​hrmagazine.co.uk. The goal is to broaden the pipeline of AI talent so domestic workers can fill new high-tech jobs, reducing the risk of unemployment. Additionally, there’s emphasis on “soft skills” like creativity, critical thinking, and adaptability – skills less likely to be automated and which enable workers to pivot into new roles. Governments and organizations like UNESCO and the OECD advocate a curriculum that blends technical skills with these cognitive and social skills to future-proof students’ careers.
• Social Safety Nets and Income Support: To cushion the transition for workers who are displaced by AI, some policy responses involve strengthening safety nets. This includes unemployment benefits, job placement services, and potentially new mechanisms like Universal Basic Income (UBI). The idea of UBI – a regular government payment to all citizens – has gained traction as a response to automation. Tech leaders (e.g. OpenAI’s CEO Sam Altman) have suggested UBI could help society manage AI-driven upheaval​urban.org. Pilot programs and studies are underway in a few places, though no major economy has implemented UBI nationwide. More immediately, countries are updating unemployment insurance and welfare programs to cover gig workers and others affected by automation. The White House (USA) in late 2023 directed the Department of Labor to examine its ability to support workers displaced by AI, including evaluating if current programs are sufficient​urban.org. This may lead to expanded or new programs to aid AI-displaced workers with both income and retraining opportunities. In Europe, discussions are ongoing about measures like reducing working hours (e.g. four-day workweeks) to spread work among more people if productivity soars with AI – effectively a way to prevent mass unemployment by work-sharing.
• Job Creation through Public Investment: Governments are also directly investing in sectors likely to create jobs. For example, many countries have infrastructure bills, green energy plans, or tech innovation funds explicitly designed to generate employment (often with an eye to absorbing workers from declining industries). The rationale is to actively create new opportunities faster than automation removes old ones. The US, through the CHIPS Act and Infrastructure Act, is investing in high-tech manufacturing and construction, creating construction jobs and semiconductor factory jobs that can hire workers from other sectors. Likewise, the UK’s industrial strategy and the EU’s recovery plans earmark funds for digital and green jobs. These policies serve as a counterweight to private-sector automation – ensuring there are alternative jobs available.
• Regulation of AI Implementation: Another approach is to manage the pace of automation. Policymakers debate measures like a “robot tax” (tax incentives or disincentives related to automation) to slow unchecked job replacement. The idea is to tax companies that replace humans with AI/robots, using the revenue to fund retraining or to discourage elimination of jobs. While not widely enacted, the concept has been floated in EU discussions and by some U.S. politicians. More commonly, governments consider setting rules for AI use in workplaces – for instance, requiring human oversight or limiting AI in roles where it might be dangerous or unethical to remove humans entirely. The EU’s proposed AI Act, while focused on safety and ethics, indirectly impacts employment by classifying high-risk AI systems (which could include those used in hiring/firing or worker management). Some officials have suggested that labor laws be updated to protect workers from arbitrary algorithmic replacement – e.g. giving employees notice or compensation if an AI system will significantly change their job or eliminate it. These regulatory approaches aim to ensure a “humane” transition, where workers aren’t simply discarded in the rush to automate.
• Encouraging AI that Augments, Not Replaces: A subtle but important policy direction is promoting AI applications that complement human labor rather than substitute it. Governments can fund research in “AI for human enhancement” – tools that make workers more productive instead of redundant. For example, AI that assists doctors (rather than automating doctors) or AI that helps customer service reps serve more people (rather than replacing the reps). By steering innovation toward augmentation, policymakers hope to preserve and even amplify human roles. Singapore’s national AI strategy, for instance, explicitly focuses on AI in partnership with humans in areas like healthcare and education, and less on pure labor-saving automation. This is more of an innovation policy stance than a labor policy, but it ultimately affects employment outcomes.

In the United States, while there is no single grand policy on AI and jobs yet, elements of the above are visible. The Biden Administration’s executive order on AI (2023) included mandates to develop an AI workforce strategy and called for reports on potential labor market impacts​

urban.org. Congress has begun discussing these issues (a House Bipartisan AI Task Force is studying AI’s economic effects​

urban.org). So far, U.S. efforts have centered on funding STEM education, expanding apprenticeship programs in tech, and leveraging community colleges for rapid re-training. For example, there are initiatives to train more workers for cybersecurity and IT roles as a way of bridging from shrinking job sectors to growing ones. Think tanks like the Bipartisan Policy Center urge a national strategy to “prepare the workforce for the jobs of the future” and emphasize inclusivity so that displaced workers in retail, manufacturing, etc., can find new careers in the AI economy​

bipartisanpolicy.org​

bipartisanpolicy.org. Discussions of more radical ideas (UBI, job guarantees) exist but have not translated into federal policy; instead, the approach is to empower workers with skills and mobility.

In the United Kingdom, the government has been proactive in acknowledging AI’s impact. Alongside the AI Action Plan’s focus on skills, the UK has also convened commissions and published white papers on the Future of Work. There is an emphasis on lifelong learning – for instance, the National Skills Fund and other programs to help mid-career workers train in new fields. The UK is also exploring strengthening its social safety net (e.g. considering how universal credit and other benefits might support those in retraining). Additionally, UK policymakers have eyed the idea of incentivizing companies to retain and retrain workers rather than lay them off – possibly through tax credits for training investments or for adopting technologies in worker-friendly ways. The British Safety Council, in a recent report, advised the UK to develop “robust regulatory frameworks” that prioritize worker safety and wellbeing in the age of AI, cautioning against a race-to-the-bottom where companies cut workers without plans for their transition​

britsafe.org​

britsafe.org. In sum, UK policy responses mix investment in skills with calls for responsible AI adoption and protections for workers.

Globally, organizations like the International Labour Organization (ILO) and OECD are guiding principles for governments. They stress the need for social dialogue – involving employers and unions in shaping the rollout of AI in workplaces. Many countries are sharing best practices on training and considering updates to labor laws. Some smaller economies are taking innovative steps: e.g. Singapore offers subsidies to companies that retrain workers for new tech roles, and Germany has well-established apprenticeship systems now being updated for digital skills. Canada and several European countries have public strategies linking AI development with job creation, ensuring that, for instance, AI research hubs also provide workforce development programs locally.

Finally, a crucial policy dimension is addressing inequality. AI’s job impacts could exacerbate income gaps (low-skill jobs more likely automated, high-skill jobs expanding). Governments are aware of this and looking at measures from raising minimum wages (to boost incomes of remaining service jobs) to supporting job growth in regions that might be hardest hit. The goal is to prevent AI-driven unemployment from concentrating in particular communities. For example, the U.S. Economic Development Administration has funded “future of work” projects in Midwest towns to diversify local economies beyond automatable industries. The UK has discussed placing new gig economy regulations to ensure AI-driven platforms (like ride-sharing, food delivery) treat workers fairly with minimum standards, so those who do lose formal jobs and turn to gig work still have protections.

In conclusion, the coming 5, 10, and 15 years will bring considerable change to labor markets in the US, UK, and globally due to AI and automation. Current unemployment rates are low, but a significant share of roles – especially in manufacturing, clerical, and routine service sectors – are at risk of being displaced. Simultaneously, growth in tech, healthcare, green energy, and other fields will create millions of new jobs. Most expert projections do not foresee skyrocketing overall unemployment if appropriate actions are taken; rather, they anticipate a transition period where many workers will need to shift occupations. The net impact of AI on employment could even be positive (net job growth) on a global scale by 2030​

businessbecause.com, though certain countries or demographics may experience higher unemployment without intervention. The key determinants will be how quickly the workforce can adapt skills, how much new economic activity AI generates, and what supportive policies are in place. With robust training programs, social support for displaced workers, and a focus on human-centric AI deployment, the US and UK can aim to maintain low unemployment even as AI transforms work. The challenge is significant – potentially millions of job transitions need to occur​

mckinsey.com – but history and research suggest that with the right measures, societies can harness AI’s benefits while minimizing long-term unemployment​

mckinsey.com. The coming decade will test our ability to proactively manage that balance.

Sources:

• Current U.S. labor force and unemployment: U.S. Bureau of Labor Statistics (2025)​bls.gov.
• Current U.K. unemployment: Office for National Statistics via House of Commons Library (2025)​commonslibrary.parliament.uk.
• Global labor force and unemployment: International Labour Organization (2023)​ilostat.ilo.org.
• Job displacement projections: McKinsey Global Institute (2017)​mckinsey.com; World Economic Forum “Future of Jobs Report 2025”​businessbecause.com; Brookings Institution (2019)​davidrousefaicp.com; PwC (2018)​pwc.com.
• Job creation projections: World Economic Forum (2025)​businessbecause.com; McKinsey Global Institute (2017)​mckinsey.com.
• Net impact assessments: McKinsey Global Institute (2017)​mckinsey.com; Tony Blair Institute (2023)​hkifoa.com​hkifoa.com; Brookings Institution (2019)​davidrousefaicp.com.
• High-risk vs growth sectors: Brookings (2019)​davidrousefaicp.com; British Safety Council/Tony Blair Institute (2024)​britsafe.org; World Economic Forum (2023)​wallyboston.com​wallyboston.com; McKinsey (2023)​mckinsey.com​mckinsey.com.
• Policy responses: Urban Institute (2023)​urban.org​urban.org; British Safety Council (2024)​britsafe.org; UK Government AI Action Plan (2024)​hrmagazine.co.uk.