I interpreted that differently: If you're going to develop technologies for humans living in cities, testing said technologies absent of human interaction woul

I interpreted that differently: If you're going to develop technologies for humans living in cities, testing said technologies absent of human interaction would likely be misleading. It's like YC advocating not talking to your users while you develop and test your product. An example they gave was self-driving cars. Are there sufficient challenges in the actions that don't involve human? It seems to me that the challenges that do involve humans are much greater in scope and importance.