The AI sector faces mounting pressure to justify massive infrastructure investments. According to consultancy Bain, the industry needs to generate $6 trillion per year by 2031 to validate the explosion in data center spending—with global annual infrastructure costs projected to reach up to $1.5 trillion by that date.
To hit these targets, AI companies must diversify revenue streams beyond traditional software licensing. Bain identifies search, advertising, autonomous robots, and self-driving vehicles as key growth areas. The consultancy also expects $1 trillion in annual revenue from enterprise productivity solutions, covering software development, marketing, and customer service automation across businesses worldwide.
The gap between current spending and projected returns highlights the stakes: without concrete, scalable business models delivering genuine value to enterprises and consumers, the industry risks a significant correction. The pressure is on vendors to move beyond experimental AI deployments and prove sustainable, profitable use cases at scale.