AI Infrastructure Investment Set to Reach $769 Billion in 2026
AI’s rapid expansion is creating a massive infrastructure build-out, with billions flowing into the computing systems, data centers and energy capacity needed to keep the technology running at scale.
Investment in artificial intelligence infrastructure and model architectures is on track to reach about $769 billion in 2026, according to McKinsey’s Technology Trends Outlook 2026.
The estimate is based on nearly $384 billion invested during the first half of 2026 and assumes the current pace continues through the rest of the year. That would put investment more than five times above the $145 billion recorded in 2025, making AI infrastructure and model architectures the largest-funded technology trend covered by McKinsey.
Data centers, chips, and power drive the spending
The investment covers the technologies needed to build and scale AI systems, including foundation models, data centres, cloud platforms, training frameworks, and data pipelines. McKinsey describes AI infrastructure and model architectures as the core technologies supporting the development and operation of AI systems.
Spending on AI infrastructure doubled in 2025 as companies expanded computing capacity. The increase is also creating greater demand for electricity and the physical infrastructure required to supply new data centres.
McKinsey said energy technologies attracted nearly $200 billion in investment in 2025, reflecting the growing importance of power infrastructure as AI computing capacity expands.
AI is attracting capital across several technology markets
AI infrastructure is not the only technology area seeing a sharp increase in investment. McKinsey’s analysis shows that five technology trends are on pace to receive more than twice as much investment in 2026 as in 2025. They are agentic software development, AI infrastructure and model architectures, AI for scientific discovery and engineering, the future of space technologies, and the future of robotics.
RELATED ARTICLES:
AI infrastructure and model architectures have the largest projected investment among those five areas, increasing from about $145 billion in 2025 to roughly $770 billion on McKinsey’s chart. The $769 billion figure reported from the same analysis reflects the underlying extrapolation.
Power is becoming a constraint
The rapid expansion of AI infrastructure is increasing pressure on electricity systems. McKinsey estimates that US data centers running AI workloads could consume as much electricity by 2030 as California does today. The firm also points to constraints involving power availability and the infrastructure needed to connect new data centers to electricity grids.
That makes electricity supply an increasingly important part of the AI investment cycle. Companies expanding computing capacity need not only chips and data-centre space, but also reliable power and the infrastructure required to deliver it.
AI investment moves deeper into physical infrastructure
The projected $769 billion highlights how AI investment is moving beyond software and into the physical systems required to support increasingly demanding workloads. The figure remains a projection for the full year. It is based on investment during the first half of 2026 and an assumption that the pace continues through the remainder of the year.
If that pace is maintained, the increase from $145 billion in 2025 would represent a substantial expansion in capital flowing into AI infrastructure and model architectures, alongside growing investment in the energy and physical infrastructure needed to support the technology.
Makon Financials
