AI technology can generate speculative-growth equilibria. These are rational but fragile: elevated valuations support rapid capital accumulation, yet persist only as long as beliefs remain coordinated. Because AI capital is labor-like, it expands effective labor and dampens the normal decline in the marginal product of capital as the capital stock grows. The gains from this expansion accrue disproportionately to capitalists, whose saving rate rises with wealth, raising aggregate saving. Building on Caballero et al (2006), I show that these features generate a funding feedback—rising capitalist wealth lowers the required return—that can produce multiple equilibria. With intermediate adjustment costs, elevated valuations are the mechanism that sustains a transition toward a high-capital equilibrium; a loss of confidence can precipitate a self-fulfilling crash and reversal.

That is from a new NBER working paper by Ricardo J. Caballero.