Artificial intelligence (AI) is no longer simply another technological revolution. It is rapidly becoming the infrastructure upon which economic power, national security, education, industry and global influence will rest. The countries that develop, control and commercialise AI will shape the twenty-first century; those that remain passive consumers will become technologically dependent, underdeveloped and economically vulnerable.
AI POLICY
Artificial intelligence (AI) is no longer simply another technological revolution. It is rapidly becoming the infrastructure upon which economic power, national security, education, industry and global influence will rest. The countries that develop, control and commercialise AI will shape the twenty-first century; those that remain passive consumers will become technologically dependent, underdeveloped and economically vulnerable.
The global race has consequently moved far beyond chatbots. AI systems can now generate software, conduct research, analyse medical images, design products, operate autonomous machines and perform complex professional tasks. Agentic AI can plan and execute multi-stage assignments with diminishing human supervision, while increasingly efficient open models are reducing the cost of adoption. Artificial General Intelligence (AGI) remains uncertain in timing, but the direction is unmistakable: machine capabilities are advancing faster than institutions and governments can adapt.
Global investment reflects this urgency. The Stanford AI Index reports that worldwide corporate investment in AI more than doubled in 2025, reaching nearly $582 billion, while private investment rose to approximately $345 billion. The United States alone attracted almost $286 billion in private AI investment and produced nearly 2,000 newly funded AI companies. China continues to accelerate technologically through efficient and increasingly powerful models, while Europe, the Gulf states and Southeast Asia are building large AI data centres, sovereign compute capacity and startup ecosystems.
AI has also become the fastest-adopted general-purpose technology in history. Generative AI reached an estimated 55 per cent of the global population within three years, faster than the mobile phone. The strategic question for Pakistan is therefore no longer whether AI will transform our economy. It is whether Pakistan will build and export AI or merely import and consume it.
Pakistan has recognised the opportunity. The National AI Policy and the government’s commitment to invest $1 billion in AI by 2030 represent a potentially historic turning point. The plan includes AI education, scholarships, infrastructure and research. The Ministry of Information Technology and Telecommunication (MoITT) has also invited proposals to establish seven National AI Hubs across Pakistan. This is an excellent and timely initiative.
However, investment alone does not guarantee economic transformation. The decisive issue is how that $1 billion is deployed. If most of it goes to additional training programmes and scholarships, Pakistan may develop more AI users and job seekers without creating enough intellectual property, exportable products, or globally competitive companies. If a significant share is channeled into AI incubators, startups, accelerators, commercialisation and venture finance, the same investment could become an engine of exports, high-value employment and GDP growth.
Pakistan is not starting from zero. It has more than 250 universities, over 2.5 million students and a large pool of computer scientists, engineers, researchers, and freelancers. Universities produce tens of thousands of research papers and potentially commercial ideas every year.
The country has also developed a considerable incubation network. HEC has established over 40 Business Incubation Centres (BICs) in public-sector universities, with additional facilities operating in private universities. The MoITT has established eight major National Incubation Centres (NICs), while the provincial governments have established their own incubators. Several private initiatives also support aspiring entrepreneurs.
This infrastructure is valuable, but its economic impact remains substantially below its potential. Pakistan has produced thousands of startup participants, pitch competitions and training cohorts, yet it has not produced a technology unicorn of global scale from within its domestic incubation system. Many university incubation centres remain administratively managed facilities rather than professionally operated venture-building organisations. Their performance is too often measured through workshops conducted, participants trained or startups admitted, not revenue generated, investment attracted, patents commercialised, jobs created, exports earned or companies successfully scaled.
The accelerator layer is even weaker. One of the few recognised private-sector accelerators supporting early-stage companies has recently closed down. Pakistan presently lacks permanent, globally connected accelerators capable of taking promising ventures emerging from BICs and NICs and transforming them into investment-ready international companies.
An incubator helps an entrepreneur develop an idea, validate a problem and build an initial product. An accelerator takes a functioning startup with market potential and rapidly helps it acquire customers, raise investment, enter international markets and scale. Pakistan has concentrated primarily on the first stage while neglecting the stage where companies become major exporters.
The proposed AI hubs can help close part of this gap, but they must not become additional buildings, training centres or disconnected research facilities. Each hub should contain a dedicated AI incubator with professional management, computing resources, access to national datasets, legal and intellectual-property support, prototype funding and strong links with industry. Their mandate must be to launch AI companies, commercialise research and solve nationally important problems.
Universities must form the foundation of this system. In the AI era, their purpose cannot remain limited to awarding degrees and publishing papers. Education must produce capability, research must produce solutions, and knowledge must produce economic and societal value. Every student should graduate not merely with a transcript but with a relevant skill, a completed project, a solved problem, a tested invention or an attempted venture.
Pakistan universities should focus their commercialisation strategy on areas where national needs intersect with export potential. This requires integrating AI across all academic disciplines, rather than confining it to computer science. Agriculture students should learn precision farming, crop monitoring, disease prediction, water optimisation, climate resilience and supply-chain management. Medical students should understand AI-assisted diagnosis and digital health. Business students should learn AI-enabled entrepreneurship, while engineers should develop intelligent manufacturing and energy systems. Financial services, education, and logistics can similarly produce scalable AI platforms for emerging markets. Universities must ultimately become laboratories of both technology and humanity, advancing innovation while preserving creativity, empathy, moral reasoning and critical thought in an age of intelligent machines.
The countries that develop, control and commercialise AI will shape the twenty-first century; those that remain passive consumers will become technologically dependent, underdeveloped and economically vulnerable.
The AI policy should become the foundation of a national AI startup production system, one capable of increasing IT exports, creating high-value employment, attracting global investment and making a measurable contribution to GDP
In addition to developing and applying AI systems to agriculture, engineering, manufacturing, health, business, finance, education and logistics, Pakistan must focus on cybersecurity, secure communications, autonomous systems, simulation, defence and national security, which all have strong strategic importance. Pakistan should develop these AI-based capabilities under strong security safeguards. Sovereign data infrastructure and affordable access to computing power will be essential, because AI startups cannot compete globally if they lack reliable cloud services, GPUs, high-speed connectivity and access to trustworthy datasets.
The emergence of efficient open-source models has created an important opening. Pakistan cannot match the hundreds of billions of dollars being invested in frontier models and hyperscale data centres by the US and China. It does not need to. It can adapt existing models, build specialised applications, develop local-language systems and create affordable solutions for markets across South Asia, Central Asia, the Middle East and Africa.
Pakistan’s greatest opportunity lies in applied AI: solving problems that wealthy technology companies may overlook but billions of people experience. Systems designed for low-resource languages, small farmers, overcrowded hospitals, informal businesses and developing-country governments could become globally exportable products. Pakistan can position itself as the AI startup hub of the Muslim world and a supplier of cost-effective solutions to emerging economies.
Pakistan has already built an economy of IT services, freelancing and outsourced labour, and its exports have risen from approximately $2.6 billion in FY2021–22 to $4.6 billion in FY2025–26, an increase of nearly 76 per cent in four years, yet the comparison with India’s technology exports exceeding $250 billion shows the scale of the unfulfilled opportunity. Pakistan’s service exports should continue growing, but a programmer or company selling hours has limited scalability.
Pakistan must now move into building scalable AI products, platforms and globally competitive companies. The strategic progression must therefore be from freelancers to teams, teams to startups, startups to scalable companies, and companies to global exporters.
This transformation will require, alongside existing initiatives, a national AI Accelerator Hub operating to international standards and governed through a public-private partnership. Its board should include accomplished technology entrepreneurs, venture capitalists, university leaders, industry experts and members of the Pakistani diaspora. Its professional management should be selected competitively and compensated on the basis of measurable performance.
The accelerator should identify the most promising ventures emerging from university BICs, NICs and the AI hubs, provide intensive acceleration, and connect them with global customers and investors. It should offer seed capital, cloud and compute credits, regulatory assistance, intellectual-property support, international mentors and market-entry programmes. Pakistani founders abroad should be encouraged to mentor, finance and co-found domestic ventures.
The government should allocate matching funds from the proposed one billion dollar AI investment that mobilises private capital rather than attempting to select every winner itself. For each public rupee invested, private investors should be encouraged to contribute additional capital. Tax incentives should reward investment in innovative companies, while regulations governing venture funds, employee stock options, foreign investment and repatriation of earnings must be simplified. Public-sector organisations can also become early customers by procuring successful Pakistani AI solutions through transparent competitive mechanisms.
Performance must be measured through outcomes: companies formed, private investment raised, paying customers acquired, intellectual property commercialised, high-value jobs created and exports earned. Incubators that consistently produce results should receive greater support; those that remain inactive should be restructured, merged or professionally re-managed.
A doable national ambition should be to increase Pakistan’s IT and AI exports to at least $12 billion by 2030. Achieving this would require sustained annual growth, hundreds of investor-ready ventures, dozens of high-growth companies and several firms capable of operating at global scale. Unicorns should not be pursued as publicity trophies, but as evidence that Pakistan can build companies whose intellectual property and markets transcend national borders.
There are legitimate concerns regarding employment displacement, surveillance, deepfakes, bias, cyberwarfare and the misuse of autonomous systems. Pakistan needs enforceable safeguards for privacy, security, transparency and human accountability. The safest position in an AI-driven world is not technological dependence but responsible national capability.
Pakistan possesses the talent, demographic scale, policy commitment and an expanding incubation network. What it lacks is the final bridge connecting knowledge with markets and startups with global scale. The AI hubs should be linked to dedicated AI incubators, existing BICs and NICs, venture financing and an AI Accelerator Hub.
The global AI economy will not wait for Pakistan’s institutions to become ready. The country must now move beyond celebrating policy announcements, training numbers and research publications. It must build products, companies, exports and wealth. The AI policy should become the foundation of a national AI startup production system, one capable of increasing IT exports, creating high-value employment, attracting global investment and making a measurable contribution to GDP.
Pakistan has trained the talent and announced the ambition. The urgent task now is to complete the ecosystem and accelerate the companies that will convert that ambition into national wealth.
The writer is a former senator and former chairperson of the Higher Education Commission.
