The technology boom of the 2010s was driven by asset-light software companies built around the SaaS model. Businesses like Salesforce, Shopify, Adobe, Netflix, and Spotify scaled globally without owning large physical infrastructure. They relied on cloud providers, outsourced logistics, and digital distribution, allowing them to grow revenue quickly while maintaining high margins and relatively low capital requirements. This model became highly attractive to investors because software could scale with minimal additional cost, creating strong operating leverage and high returns on invested capital. Watch the video below to understand how AI is changing this entire model.
The rise of generative AI has shifted the economics of technology. Unlike traditional SaaS applications, AI systems require massive physical infrastructure. Training and running large AI models demand thousands of GPUs, advanced semiconductors, high-density data centers, storage systems, cooling technology, and enormous amounts of electricity. This has created pressure across the entire infrastructure stack, including chips, networking, power grids, and data center capacity. Major technology companies such as Amazon, Microsoft, Alphabet, and Meta Platforms are now spending heavily on AI infrastructure to support future demand.
As a result, asset-heavy businesses have regained strategic importance. NVIDIA became central to AI deployment due to its dominance in GPU supply, while hyperscalers benefit from owning cloud and data-center infrastructure that is difficult to replicate. Demand has also expanded toward storage, networking, and power infrastructure providers. Utilities have re-entered growth discussions as AI data centers require a stable electricity supply, leading companies like Meta and Microsoft to secure long-term energy agreements.
However, this new cycle also introduces risks. Heavy infrastructure spending can reduce free cash flow and create oversupply if capacity grows faster than demand. Historical examples, such as the telecom infrastructure boom of the 1990s, show how excessive investment can pressure margins and destroy capital. Even so, those infrastructure systems later became foundational to the internet economy. Similarly, today’s AI infrastructure—chips, storage, data centers, and power networks—is increasingly becoming the utility layer of the next digital era.
Disclaimer: Investments in securities markets are subject to market risks. Read all the related documents carefully before investing. The securities quoted are exemplary and are not recommendatory.
