The Price-to-Dream Ratio
Autonomous shopping agents, co-working and research agents, and coding agents that write, test, and deploy software. The demos are impressive and the announcements relentless, but how much economic value is any of this generating? Almost all AI-related spending is capital expenditure. Companies are buying chips, building data centers, and scaling up cloud capacity. This is spending on AI infrastructure, not productivity from AI deployment. AI is in the infrastructure buildout phase, not the value capture phase. Mass spending is generating minimal returns, but the market has decided to price the dream rather than the earnings. Can any of this translate into economic reality before the capital runs out and political patience expires? Infrastructure, capability, and revenue growth are happening. Productivity is developing. But a significant gap still exists between capital investment and return on that investment. AI is risky, but these investments are not irrational. They are pricing the dream, and the long-term winners remain unclear.