ARTIFICIAL intelligence is usually presented as weightless, almost entirely digital. We ask it to write, calculate, translate or create an image, and the response appears instantly. Yet every answer depends on a substantial physical system. This comprises data centres, specialised processors, cooling equipment etc.
In reality AI has both ‘brain’ and ‘muscle’. Algorithms, software, researchers, engineers and companies comprise its brain. Its muscle consists of electricity, computing infrastructure, minerals, manufacturing capacity and reliable networks. The first phase of global AI competition focused mainly on the brain. As AI becomes more widely used, the muscle is becoming impossible to ignore.
The International Energy Agency expects electricity demand from data centres to more than double by 2030, with AI a major driver. The challenge is not simply producing more power. It is producing reliable electricity supply and expanded transmission networks to accommodate the needs of large data centres. Oil and gas reserves beneath the ground are not the same as electricity delivered to a data centre. Resources become strategic only when they can be converted into dependable infrastructure.
The physical foundations of AI also include copper, aluminium, silicon, and rare-earth materials. Advanced semiconductor technologies depend on specialised materials, while mining and refining remain concentrated in a limited number of countries. The digital economy has not escaped geography! An algorithm can cross a border in milliseconds, but a power plant, mine or factory cannot.
This creates a new geopolitical equation. Countries with the strongest AI research and software ecosystems may still depend on others for energy, minerals and manufacturing. Resource-rich countries, meanwhile, may possess the muscle of the AI economy without the universities, capital, software and semiconductor capabilities needed to turn it into computing power. The long-term advantage may belong to countries that can connect the two.
The digital economy has not escaped geography.
Pakistan illustrates this challenge clearly. It is not currently a leading frontier-AI power and faces serious constraints in reliable electricity, advanced computing, research funding etc. Yet it has a large young population, growing software and IT-services sectors, universities, engineering talent, hydropower and renewable-energy potential, and significant mineral resources. The strategic challenge is therefore not simply to acquire more resources or develop more algorithms, but to connect the country’s physical assets with its scientific and technological capabilities.
That would require moving beyond traditional models of exporting raw materials and low-value services. An AI-era strategy would use reliable energy, connectivity and mineral resources to support data centres, cloud services, local research, advanced applications and value-added manufacturing. AI could strengthen sectors such as agriculture, textiles, logistics, mining, health, education and public administration. In this model, AI becomes an industrial multiplier rather than merely another software sector.
Resource wealth alone, however, guarantees neither technological independence nor prosperity. Natural resources cannot substitute for universities, skilled workers, capital, innovation and capable institutions. The relationship between AI’s brain and muscle is better understood as multiplication rather than addition. A country with excellent algorithms but unreliable power will struggle to scale, while a country with abundant energy but weak research capacity may remain a commodity supplier.
The lesson extends beyond Pakistan. Countries such as the United States have extraordinary strengths in research, capital, technology companies and advanced computing. China combines rapidly advancing AI with industrial scale and manufacturing capacity. Gulf countries possess energy and capital, while Australia and Canada have substantial mineral and energy advantages.
None controls every part of the AI system. Interdependence will remain unavoidable, but countries that connect more stages, from energy and minerals to computing and applications, will have greater strategic influence.
For Pakistan and other natural resource-rich countries, the message is immediate. An AI strategy cannot stop at applications, start-ups or policy statements. It must also address reliable electricity, data-centre infrastructure, digital skills, research institutions, mineral value addition and industrial capacity.
The next phase of the AI race will not be decided solely by who builds the smartest models. It will also be decided by who can power them, supply them and apply them productively. AI has a brain. But without muscle, even the most brilliant brain cannot move.
The writer is an educationist.
Published in Dawn, October 28th, 2026
