Rising Component Costs Risk Creating Global 'AI Divide', GSMA Warns

A new global "AI divide" is emerging as the cost of essential smartphone hardware skyrockets, threatening to exclude billions of people from the digital economy. The GSMA warns that while AI technology advances rapidly, the hardware required to access it is becoming unaffordable for those in low- and middle-income countries.

According to the GSMA, the crisis is driven by a supply chain shift where memory and chipsets are being diverted toward AI infrastructure and data centers. This has led to a dramatic price surge: memory costs more than doubled between the third quarter of 2025 and the first quarter of 2026, followed by an additional 80% to 90% increase in the second quarter of 2026. The impact is most severe in the budget sector, with global smartphone shipments forecast to suffer their largest annual decline on record due to the collapse of the sub-$100 handset segment.

The Usage Gap

This hardware inflation exacerbates the existing "usage gap"—a phenomenon where people live within mobile broadband coverage but lack the devices or skills to go online. Currently, approximately 3.4 billion people do not use mobile internet, despite more than 90% of them living within coverage areas.

For the world's most vulnerable, the barrier is purely financial. By the end of 2025, an entry-level internet-enabled handset cost the poorest 20% of people in Sub-Saharan Africa 76% of their average monthly income. While mobile internet adoption has reached 4.8 billion users, the pace of growth is slowing, with only 160 million new users added in 2025 compared to 190 million in 2024.

Economic and Social Stakes

The consequences extend beyond simple connectivity. As healthcare, education, and financial services migrate to AI-enabled platforms, the lack of affordable hardware creates a systemic barrier to essential services. This shift risks permanently excluding billions from the productivity gains associated with AI, deepening existing global inequality.

"The greatest risk is not simply an AI divide between countries, but between people who can afford to participate in the digital economy and those who cannot," said Vivek Badrinath, Director General of the GSMA.

The Cost of Exclusion

Beyond the social impact, the economic cost of this divide is staggering. The GSMA estimates that closing the mobile usage gap could generate $3.5 trillion in additional GDP between 2023 and 2030. Over 90% of these economic gains would benefit low- and middle-income countries, provided the hardware becomes accessible.

What remains to be seen is whether manufacturers will develop lower-cost AI-compatible hardware or if the diversion of components to data centers will continue to cannibalize the consumer budget market. For now, the trajectory suggests that without intervention, the AI revolution may only be available to the world's wealthiest populations.