AI Could Reshape 61% of Pakistan’s Service Jobs. But “Exposed” Doesn’t Mean “Doomed”

AI Could Reshape 61% of Pakistan’s Service Jobs. But “Exposed” Doesn’t Mean “Doomed”

A new World Bank report has produced a headline number that understandably alarms a lot of Pakistanis: artificial intelligence could reshape 61% of the country’s service-sector jobs. That is a huge figure, covering millions of workers in shops, banks, offices, schools, and more. It sounds like a warning that most service jobs are about to vanish. But that reading, while natural, is wrong, and getting the nuance right matters enormously for how Pakistanis should respond.

Because here is the crucial truth buried beneath the scary headline: “high AI exposure” does not mean “about to lose your job.” The same report that flagged the 61% figure is careful to distinguish between AI that replaces workers and AI that helps them do their jobs better, and the difference is everything. So what does the 61% actually mean, which jobs should genuinely worry, and what should Pakistan’s service workers do? Here is the honest, nuanced picture, neither alarmist nor dismissive.

What the World Bank Actually Found

Let’s start with the data, accurately, because precision matters here. The World Bank report, titled “Job Opportunities and AI in Pakistan,” analyzed how exposed different Pakistani jobs are to artificial intelligence. Its headline finding: 61% of workers in Pakistan’s service sector are employed in occupations with above-median AI exposure.

To put that in context, the service sector is by far the most exposed. The corresponding figures are just 7% in manufacturing, around 5% in other industrial sectors, and less than 1% in agriculture. So AI’s workforce impact in Pakistan is heavily concentrated in services. Within that sector, some areas are especially exposed: more than 80% of workers in wholesale and retail trade, finance, insurance, real estate, and education hold jobs with high AI exposure. Wholesale and retail trade in particular stands out, because of its large size, as a key area that will shape how AI affects Pakistani employment overall. So the finding is real and significant: a large majority of Pakistan’s service workers are in roles that AI will touch. But, and this is the heart of the matter, “touch” and “exposure” are not the same as “replace.” The report is explicit about that distinction, and it changes everything.

The Crucial Distinction: Exposed Is Not Replaced

Here is the single most important point in this entire discussion, and it is the one the scary headlines miss. “AI exposure” means a job could be significantly affected by AI, but that effect can go two very different ways: AI can substitute for the worker (replace them), OR AI can complement the worker (help them do their job better and faster). High exposure tells you AI will matter for that job; it does NOT tell you whether the worker is helped or replaced.

The World Bank report makes this distinction central, and the numbers are genuinely reassuring when you look closely. It separates highly-exposed jobs into two types: those where AI mainly helps the worker (high complementarity), and those where AI could mainly replace the worker (low complementarity). And the breakdown matters enormously: around 18% of Pakistan’s employed workers are in jobs with both high AI exposure AND high human-AI complementarity, meaning AI is likely to help them, making them more productive and valuable, not replace them. Meanwhile, only about 8% of workers are in highly-exposed roles with low complementarity, the jobs facing genuine replacement risk. (The report, based on 2024 data, estimates roughly 8% of all jobs are at risk of being replaced by AI, a number that may rise, but is far from the headline 61%.) So the honest picture is this: yes, 61% of service jobs are exposed, but for a large share of them, AI is more likely to be a helpful tool than a replacement. Exposure is not doom. That distinction is the key to everything that follows.

Which Jobs Should Actually Worry (and Which Shouldn’t)

Let’s make this practical, because the real question most people have is: “Should I worry?” The answer depends on which side of the exposure line your job falls, and understanding that helps you respond.

Broadly: jobs at higher replacement risk tend to be routine, repetitive, and rules-based, work that AI can fully automate, like basic data entry, simple repetitive clerical tasks, routine form-processing, and other highly standardized work. If your job is mostly doing the same predictable tasks over and over, it is more vulnerable. Jobs where AI is more likely to help (complement) tend to involve human judgment, relationships, creativity, complex problem-solving, and skills that pair well with AI, work where AI handles the routine parts while you add the human value. A banker who uses AI to analyze data faster but still advises clients, a teacher who uses AI tools but still mentors students, a retail manager who uses AI for inventory but still leads people and serves customers, these are enhanced by AI, not replaced. The practical takeaway: whether you should worry depends less on your industry (most service workers are “exposed”) and more on what you actually do and whether you use AI to your advantage. Routine, replaceable tasks are the risk; human judgment, relationships, and AI-augmented skills are the safety. And crucially, many workers can move themselves from the first category to the second, which is the whole point.

Why This Is a Choice, Not a Fate

Here is the genuinely empowering insight, and it is the reason this is a Hot Take worth sitting with: for most “exposed” Pakistani workers, whether AI becomes a threat or a tool is largely within their control. The 61% figure sounds like a verdict, but it is really a fork in the road.

The World Bank itself stresses that AI’s actual impact will depend on how businesses and workers adopt the technology, on skills, investment, and choices, not on some fixed, inevitable outcome. This matters enormously. It means an “exposed” worker is not passively waiting to be automated; they have real agency. The worker who learns to use AI tools, who moves toward the judgment-and-relationship parts of their job, who upskills, turns their “exposure” into an advantage, becoming the AI-augmented professional who is more productive and valuable than before. The worker who ignores AI, stays stuck in routine tasks, and doesn’t adapt is the one genuinely at risk. In other words, exposure is the situation; your response determines the outcome. This reframes the whole 61% figure from a prophecy of doom into a call to action. The technology is coming to service work regardless; the question is whether each worker (and Pakistan’s workforce as a whole) adapts to ride it or gets left behind by it. That is a choice, individually and nationally, and choices can be made well.

The Honest Risk: It’s Not All Rosy

In keeping with honest analysis, let’s be clear-eyed, because this is genuinely serious, not something to wave away. While “exposed is not doomed” is the crucial nuance, real risks remain and must be acknowledged.

The honest concerns: the roughly 8% in high-replacement-risk roles face genuine danger, and that number could rise as AI improves, representing real people whose livelihoods are threatened. Adapting (learning AI, upskilling, moving to higher-value work) is easier said than done, many workers lack the resources, access, or support to do it, especially in a country with the skills gaps we have discussed. There is a real danger of a widening divide: those who adapt and benefit from AI’s productivity gains pull ahead, while those who cannot are left behind, potentially pushed into lower-skill, lower-paying work, worsening inequality (a concern the report and commentators raise). Pakistan’s broader context makes it harder: as the World Bank has noted elsewhere, countries with weak formal-job creation and high youth unemployment (like Pakistan) are more vulnerable to AI disruption. And without serious preparation (skills, support, good policy), the optimistic “everyone just adapts” scenario won’t materialize on its own. So the nuance (“exposed is not doomed”) is true and important, but it is not a reason for complacency. The positive outcome (AI as a tool that boosts productivity and workers) is possible but not automatic; it requires real effort, adaptation, and support. Both the opportunity and the risk are genuine.

What Pakistan and Its Workers Should Do

Given all this, here is the constructive path, because the right response turns the 61% risk into an opportunity. The actions follow directly from the analysis.

For individual workers: learn to use AI tools in your field (become the AI-augmented professional, not the one competing against AI); focus on the human-value parts of your job (judgment, relationships, creativity, problem-solving, service) that AI complements rather than replaces; move away from purely routine, repetitive tasks; and keep upskilling toward higher-value, less-automatable work. For businesses: adopt AI to boost productivity, but do so in ways that augment and upskill workers rather than simply replacing them, and invest in training your people. For the government and institutions: this is urgent, prepare the workforce through large-scale reskilling and AI-literacy programs (connecting to the skills-crisis fixes we have discussed); support workers in transition; invest in the skills and infrastructure that let people adapt; and craft policy that steers AI toward augmenting workers and broad benefit, not just displacement and concentration. The unifying goal is to help Pakistan’s huge service workforce move from the “at-risk” side of AI exposure to the “augmented and more valuable” side, individually, in businesses, and nationally. The 61% exposure is the challenge; turning most of it into human-AI collaboration rather than replacement is the opportunity, and the work ahead.

Industry Impact: Why This Matters for Pakistan

This finding sits at the heart of Pakistan’s economic future.

For the service workforce, the sector employs a huge share of Pakistanis, so how AI reshapes it (helping or replacing) will profoundly affect millions of livelihoods and the economy.

For productivity and growth, if AI augments service workers well, it could meaningfully boost Pakistan’s productivity and growth, a genuine opportunity the World Bank highlights.

For inequality, the outcome could narrow or widen divides, depending on whether Pakistanis broadly adapt and benefit, or whether gains concentrate among a few, making inclusive preparation crucial.

For the skills imperative, this reinforces the urgency (echoing Pakistan’s broader skills crisis) of reskilling and AI-literacy, the single biggest factor in whether AI exposure becomes opportunity or disruption for Pakistani workers.

Expert Insight: The 61% Is a Fork in the Road, Not a Verdict

The essential insight is that the World Bank’s “61% of service jobs exposed to AI” finding should be read not as a doom-laden verdict but as a fork in the road, a clear signal that AI will transform Pakistan’s service work, with the direction (augmentation or replacement) depending heavily on how workers, businesses, and the country respond. The nuance is genuinely crucial and genuinely hopeful: high exposure does not mean job loss. The report’s own breakdown shows that far more workers (around 18%) are in jobs where AI is likely to help them than in jobs facing genuine replacement (around 8%). For most exposed workers, AI is more plausibly a powerful tool than a pink slip, if they learn to use it and lean into the human parts of their work. That reframes the whole conversation from passive fear to active agency.

But the honest wisdom pairs that hope with realism and urgency. The positive outcome is not automatic. The roughly 8% at real replacement risk face genuine danger; adaptation is hard and unevenly accessible; and without serious reskilling, support, and good policy, AI could widen Pakistan’s inequalities and strain its already-fragile job market, exactly as the World Bank warns for vulnerable economies. So the task is clear and dual: individually, Pakistani service workers should treat this as a wake-up call to learn AI, move toward judgment-and-relationship work, and upskill, turning their exposure into an advantage. Nationally, Pakistan must urgently prepare its workforce through large-scale reskilling, AI literacy, worker support, and policy that steers AI toward augmenting people and broad benefit. Get this right, and the 61% becomes a story of a more productive, AI-empowered Pakistani workforce. Get it wrong, or do nothing, and it becomes a story of disruption and widening divides. The window, as the World Bank puts it, is narrow, but the choice is real. “Exposed” is not “doomed,” but it is a call to act. Pakistan’s service workers are not passengers in this; they can be drivers, if they, and the country, choose to adapt now. That is the honest, empowering bottom line.

Future Outlook

Expect AI to increasingly reshape Pakistan’s service work over the coming years, with the real-world impact unfolding gradually and depending heavily on adoption, skills, and policy. Watch which way the balance tips: toward widespread human-AI collaboration (workers augmented and more productive) or toward displacement (especially for the most routine roles). Watch too whether Pakistan mounts the serious reskilling and support efforts needed to steer toward the better outcome.

If workers adapt and Pakistan prepares its workforce well, the 61% exposure could largely translate into productivity gains and augmented, more valuable jobs, a genuinely positive transformation. If adaptation lags and preparation is weak, more of that exposure could become displacement and inequality. The trajectory is not fixed; it depends on choices being made now, by workers, businesses, and the government. The encouraging reality is that the path to the good outcome is known (adapt, upskill, augment, support); the task is to walk it urgently. For Pakistan’s service workforce and economy, how this is handled over the next few years will be genuinely consequential.

Conclusion

The World Bank’s finding that 61% of Pakistan’s service-sector jobs face high AI exposure is a serious, important signal, but it is not the doom-laden verdict the headline suggests. The crucial, hopeful nuance is that “exposed” does not mean “replaced”: AI can either substitute for workers or help them, and the report’s own data shows far more Pakistani workers (around 18%) are in jobs where AI is likely to assist them than in jobs facing genuine replacement (around 8%). For most exposed service workers, AI is more plausibly a powerful tool than a threat, if they learn to use it and lean into the human-value parts of their work: judgment, relationships, creativity, and service. That reframes the 61% from a prophecy into a fork in the road, where the direction depends on how workers, businesses, and Pakistan respond. The honest caveats are real: the roughly 8% at genuine replacement risk face real danger, adaptation is hard and unevenly accessible, and without serious reskilling and support, AI could widen Pakistan’s divides. So the message is both empowering and urgent: individually, Pakistani service workers should treat this as a wake-up call to learn AI, upskill, and move toward augmented, higher-value work. Nationally, Pakistan must urgently prepare its workforce. Do that, and the 61% becomes a more productive, AI-empowered future. Ignore it, and it becomes disruption. “Exposed” is not “doomed”, but it is a call to act, now. The choice, genuinely, is Pakistan’s to make.

This article is an analysis and opinion piece for general informational purposes, based on the World Bank report “Job Opportunities and AI in Pakistan” and related reporting available in 2026. Figures (including the 61% exposure, 18% complementarity, and 8% replacement-risk estimates) are as reported by cited sources and based on the World Bank’s methodology and 2024 data; they may vary and evolve. Individual outcomes depend on many factors. This is not career, financial, or policy advice.