Artificial intelligence is no longer a topic that belongs only to laboratories and technology companies, because it has quietly become an economic force that is changing how goods are made, how workers are hired, where investors put their money and how nations compete with one another, and the uncomfortable question that follows is whether this force will bring the world closer together or push it further apart. The International Monetary Fund said in Bangkok this week that money flowing into AI is helping to keep global growth near three percent, which is welcome news in a year of energy shocks, but a healthy global number can easily hide an unhealthy distribution, and the evidence suggests that the benefits of this technology are not being shared in anything like equal measure.
The clearest sign of this is the way the building of AI is concentrated in a small group of rich countries. According to the World Bank’s Digital Progress and Trends Report 2025, high-income countries produced 87 percent of the notable AI models, 86 percent of the AI start-ups and 91 percent of the venture capital funding, even though they hold only 17 percent of the world’s population, and these figures matter because whoever builds the technology also collects the patents, the high-paying jobs and the power to set the rules of the digital economy. Even within Asia, which is gaining from the boom, much of the gain is going to countries that supply chips, components and data-centre equipment, which is useful business, but a supplier of hardware should not mistake a full order book for real control over the technology that the hardware serves.
Behind this concentration lies a gap in basic infrastructure that is easy to overlook, since AI depends not only on clever software but also on steady electricity, affordable internet, powerful computers and data centres. The World Bank reports that 93 percent of people in high-income countries use the internet, compared with 54 percent in lower-middle-income countries and only 27 percent in low-income countries, so a student in a well-connected country can use AI tools to learn coding or improve her writing every day, while a student without a reliable connection or a suitable device is left outside the door. In this way the old digital divide becomes the new AI divide, and the people who were already behind find that the race has become faster.
Skills and education add another layer to the problem, because AI does not make a worker productive on its own, and a person needs the knowledge to use it well, to check what it produces and to apply it to real problems. Countries with strong universities, technical training and habits of lifelong learning will turn AI into growth much more easily, while countries that fail to prepare their workforce may find that their people can open the applications but cannot create the high-value work that sits behind them. The United Nations Development Programme warned in its report “The Next Great Divergence” that the half-century of poor countries slowly catching up with rich ones could now be reversed, and it found that jobs held by women are almost twice as exposed to automation and that employment among young people aged twenty-two to twenty-five is already falling in the occupations most touched by AI. The President of Singapore, Tharman Shanmugaratnam, made the same point when he warned that demand for entry-level workers is shrinking, and a first job matters far more than its salary, because it is where a young person learns how work actually happens.
For Pakistan, all of this is close to home, since we have one of the youngest populations in the world and a large body of freelancers and IT workers who earn foreign income through the internet. AI can help our students learn, our software developers work faster and our freelancers serve international clients, but the same tools can also take over the routine digital tasks on which many of these workers depend, and unequal access to devices, internet and practical training could widen the gap between students from different backgrounds. Using AI is not the same as having the capacity to earn from it, and if we only consume foreign platforms and cloud services we will remain dependent on companies abroad for prices, data rules and even the continued availability of the services we rely on, which is not an argument against foreign technology but an argument for having enough local skill and bargaining power to use it on fair terms.
Yet it would be wrong to say that AI must make poor countries poorer, because the technology can also lower some of the old barriers to growth. A small shopkeeper can now get customer service, accounting and marketing help that once required expensive software or a trained staff, a teacher can prepare lessons in less time, and a farmer can learn about crop disease, irrigation and market prices at a very low cost. The World Bank president, Ajay Banga, spoke in Bangkok about smaller and cheaper AI tools that do not depend on giant models and can still raise productivity in agriculture, health and education, and this is a far more realistic path for Pakistan than trying to compete with America and China in building enormous systems. In a country where specialist doctors are few and many farmers have little advice, even modest tools, if they are reliable and used under proper professional supervision, can make a real difference.
The divide, then, is not fixed, and what happens next depends on the choices that governments, businesses, universities and international partners make now. Pakistan needs to expand affordable broadband and reliable electricity, teach digital skills and critical thinking from school to university, and support local start-ups that build tools for local problems, including tools that work well in Urdu and our regional languages, which foreign companies are unlikely to treat as a priority. Small businesses need training and financing so that the gains do not stay with a few large firms, the state needs sensible rules to protect personal data and to deal with bias and accountability, and our policymakers should also be thinking about how a shift from labour to capital will affect the tax they collect from salaries.
The leaders at Bangkok showed that they understand the risk, but understanding costs very little while action costs a great deal, and the countries that do well in the years ahead will be those that invested early in their people, their infrastructure and their own ability to innovate. If Pakistan waits, it will not be AI that leaves us behind, it will be our own delay, and so the sensible course is to begin now, while the gap is still small enough to close.
Read morwe: https://digdebate.com/ai-policies-universities-assignment-rules/
Our Editorial Desk is the intellectual engine of Digital Debate, responsible for the rigorous research that anchors every conversation. Our team deep-dives into data, checks every source, and consults academic literature to move beyond headlines and identify the questions behind the questions.
