COMMENT: How technology could shape our next decade

LAHORE: The technological landscape in 2025 is not just ‘smarter’ phones and cheaper servers — it is an economy being remade by artificial intelligence (AI) and related frontier technologies. While Pakistan has taken quite a few early steps, the scale and speed of change is very slow.

Though universities are producing AI graduates, the government has moved from strategy drafts to a national policy. After years of consultations, Pakistan released its National AI Policy in 2025 that sets out objectives on skills, research, data governance and public-sector adoption. This is an essential foundation, but a policy on paper must be followed by budgets, measurable targets and institutional capacity to execute.

Pakistan’s higher-education system expanded connectivity and course offerings in the past five years — dozens of universities now list AI/ML programmes and the Higher Education Commission has broadened research networks — producing more graduates with basic AI skills. Yet quantity masks uneven quality: advanced research groups, high-end compute facilities and PhD pipelines remain thin relative to needs.

A domestic AI services and product ecosystem has emerged — systems integrators, boutique AI consultancies and a growing list of startups offering computer-vision, NLP and analytics services — but most firms are small, focused on services/exportable engineering work, and depend on foreign models, cloud platforms and talent. Venture capital has increased but remains modest compared with regional neighbours.

Studies of Pakistan’s labour market show meaningful exposure to automation: a sizable share of routine tasks is susceptible to AI-driven change. Some analyses estimate a non-trivial proportion of jobs at high risk or likely to change substantially within a few years — a pattern that will intensify as generative AI spreads across white-collar as well as blue-collar work. That prospect reinforces the imperative for reskilling, stronger social safety nets and active labour-market policies.

India has leveraged scale — a large domestic tech sector, massive IT services exports, government programmes (IndiaAI, NITI Aayog strategy) and major corporate investments — to move rapidly into productisation, generative-AI labs and national planning for AI-driven growth. Indian conglomerates and startups are investing heavily in GenAI, enterprise AI and public-good applications, and India’s policy and industrial push have raised it well above Pakistan in both capacity and ambition.

Bangladesh and Sri Lanka have also accelerated AI planning and capacity-building. Bangladesh has prioritised an AI readiness strategy and growing startup activity; Sri Lanka has a national AI strategy focused on governance and skills. Their ecosystems are smaller than India’s but — critically — both have concentrated public efforts to skill people and pilot AI in government services, narrowing some capability gaps at the application layer even if deep research capacity remains limited.

Pakistan has made some progress like policy commitments, more degree programs, an embryonic startup scene, but it lags India on scale and productization, lags Bangladesh and Sri Lanka in execution of national pilots and skilling in some areas, and is far behind China in core AI infrastructure. Closing these gaps will not be easy, but the alternative is sustained dependency and missed opportunities.

Planners must use government demand in health, agriculture, citizen services and revenue administration to create markets for local AI solutions and to validate responsible deployment models. Procurement rules should require explainability, privacy safeguards and local capacity-building.

Establish clear, rights-respecting data frameworks that enable safe data sharing for research (with privacy safeguards), coupled with investments in secure digital identity and interoperable data infrastructure.

There must be incentives for sectors like textiles, agri-value chains, logistics, and fintech with matching grants, tax credits for R&D, and focused acceleration programs that pair firms with research labs.

AI will shift profit pools, displace tasks and create new kinds of work. But the net outcome is not preordained: it depends on policy choices. Pakistan’s 2025 policy and expanding educational pipeline are good first steps, but ambition must match urgency. Planners who invest in computers, skills, governance and catalytic public procurement will give citizens a fighting chance to prosper as the world adopts these powerful technologies.