Enterprise software just got its verdict. $285 billion wiped from valuations, multiples compressed, and the assumption underneath thirty years of software design — that humans navigate the stack — is now breakable. This isn't a market correction. It's an architectural reset. Drawing on work inside hundreds of enterprise AI programs, this session maps what the reset actually looks like from the frontlines: the data foundations that have to be rebuilt before agents can reason reliably, the workflow boundaries where most AI investments quietly die, the governance gaps that emerge when agents move from tools to labor, and the trust architecture that determines what actually gets deployed at scale versus what stalls in the boardroom. The enterprises that win won't be the ones that deploy the most agents. They'll be the ones that govern them best — and that starts with understanding what the new stack actually requires of the humans inside it.