AI cancer treatment collides with chip shortage: why does the bet on clinical data outpace supercomputing?

Renée Haas, CEO of Arm Holdings, believes that artificial intelligence can arrive at cancer treatments that human minds cannot achieve within our current generations, confirming that simulating the disease's impact on genetic markers and cells represents an extremely complex problem that exceeds the power of contemporary computers and human capabilities. However, this promising medical horizon collides directly with a sharp shortage of electronic chip supply, which hampers the rapid expansion of building vital data centers to run these complex mathematical models.

Arm, which is based in Cambridge and holds the highest market valuation in the history of British companies, has become a core pillar in the architecture of low-power processors, wherehalf of the world's artificial-intelligence data centers rely on its designsThis presence prompted the company to sell its own chips and develop the Arm AGI processor for Meta, recording orders exceeding two billion dollars since its launch in March, amid a global environment suffering from supply-chain constraints and manufacturing concentration at Taiwan's TSMC.

In contrast, Professor Chris Bacall, from the London Cancer Research Institute and CEO of Sentinel4D, puts forward a parallel view that the success criterion is not tied to the size of computers but to the quality of input data. He explained that his lab trains models on precise measurements extracted from the patients' own samples rather than data aggregated from the internet, noting thatthe future of medical AI belongs to those who possess precise clinical measurementswhich can shave years off drug-development pathways without the need for massive data centers.

This hardware bottleneck extends to plans for building data centers with gigawatt-scale capacity in the United States, France and even orbital proposals in space, and it also casts shadows over the spread of humanoid robots, which Haas expects to see widely deployed within five years to perform inspection, maintenance and service tasks with continuously self-learning capabilities. According to his view, the idea of locating advanced chip factories in countries such as Britain remains unfeasible, given the exorbitant costs, the need for natural resources, scarce specialised expertise and a fully integrated environmental system.

This scenario carries direct strategic implications for the health and technology sectors in the Gulf region, Egypt and the Levant, as it compels medical and academic institutions to shift their investments from the race to secure massive, costly computing capacity toward building ultra-high-quality local genomic and clinical databases, a path that gives research teams a genuine competitive edge in developing personalized medicine. It also requires infrastructure managers and engineers in the region to redirect data-center plans toward energy-efficient architectures to lower operating costs, while hedging procurement schedules against delays in advanced-hardware supply chains.