there is a faster-horse problem in AI that seems to be infiltrating the venture capital and startup ecosystem discourses rapidly. it is opportunistic at best and myopic at worst. when it comes to AI, I think we should make fewer declarative statements and ask more questions. I think we should be students, not teachers. when we frame conversations in terms of how existing systems or processes can be "supercharged" or "made more efficient" by AI, we fail to leave room for the imagination – the creative solutions that rearchitect the status quo altogether. we are asking for faster horses versus the car. these are modest transformations. generational AI companies will instead focus on questions like the following: what new and emergent behaviors are unlocked by large-scale models, and how might these behaviors reimagine the design of these systems at large?