admission from one of the world’s leading AI hardware firms.
Research backs the financial reality
A 2024 MIT study adds weight to this view. Researchers analysed the economic feasibility of replacing human workers with AI across visually intensive roles and found automation only made financial sense in around 23% of those jobs. In the remaining 77%, keeping human workers was still the cheaper option. Experts note that AI will only become truly cost-effective once it is both cheaper to run and more reliable — requiring less human supervision to function correctly at scale.
Big tech keeps spending anyway
Despite the questionable economics, investment in AI continues to surge. Morgan Stanley estimates tech firms have already committed around $740 billion to AI-related spending this year alone — a 69% increase compared to 2025. McKinsey projects that total AI spending could reach $5.2 trillion by 2033, covering data centres, IT infrastructure, and software. Meanwhile, more than 92,000 tech workers have been laid off so far in 2026, a pace that outstrips all of last year’s redundancies combined. Analysts describe the current situation as a short-term mismatch between financial logic and corporate ambition — with companies betting heavily that today’s expensive AI will become tomorrow’s competitive edge.