A recent study by Bain & Company forecasts a growth trajectory for the artificial intelligence (AI) market, predicting an annual increase of 40% to 55%. By 2027, the AI-related hardware and software market is expected to reach between $780 billion and $990 billion.

According to Bain & Company’s Technology Report 2024, this expansion is fueled by three primary factors: the development of larger AI models and data centers, the rise of enterprise and sovereign AI initiatives, and improved software efficiency. These elements are poised to propel the sector toward a near trillion-dollar valuation within the next three years.

AI workloads are projected to increase by 25% to 35% annually through 2027, driving a substantial rise in computing power demand and accelerating the growth of large-scale data centers.

Bain & Company estimates that current data centers, which typically operate between 50 and 200 megawatts, will need to expand to over a gigawatt in the next decade. Consequently, the cost of establishing these facilities could escalate from the current range of $1 billion to $4 billion to as much as $10 billion to $25 billion over the next five years.

This escalation will transform the ecosystems surrounding data centers, affecting infrastructure engineering, power production, and cooling systems, while putting further pressure on supply chains.

The rising AI-driven demand for graphics processing units (GPUs) could increase the overall demand for specific upstream components by over 30% by 2026. Like the surge in PC demand during the pandemic, the growing need for AI computing power will likely strain supply chains for data center chips, personal computers, and smartphones, potentially ushering in a new wave of semiconductor shortages.

Bain also highlights that emerging “sovereign” AI blocs will introduce additional complexity for technology companies. Countries like Canada, France, India, Japan, and the UAE are significantly investing in their domestic AI capabilities, channeling billions into developing local computing infrastructure and training AI models on homegrown data.

This strategy aims to enhance control over their digital ecosystems, potentially reshaping global AI development and creating new opportunities and challenges for industry players.