[AINews] Open Models, Model Labs vs Agent Labs, and What's Untrainable — Sarah Guo
Sarah Guo argues AI apps earn defensibility in the 'untrainable' zone by arranging a company's private reality so models can act on it.
“An application earns its place in the untrainable corner by doing unglamorous work: arranging a company's private reality so a model can act on it, handing the model the tools to act, working with the customer to change the reality of its workforce.”— Sarah Guo
VC Sarah Guo lays out a framework distinguishing model labs from agent labs, arguing that durable application moats come from the 'untrainable' work of translating a customer's private context and workflows into something a model can act on. It matters because it reframes where startup defensibility lives as foundation models commoditize, favoring companies doing unglamorous integration over those chasing raw model capability.