"Natural language is actually a very difficult thing to emulate," Nedim Fresko, Amazon's VP of Alexa Devices and Developer Technologies, told ZDNet last year. "When people speak naturally, they change direction, they make contextual references to things they said. Sometimes they over-supply information, sometimes they under-supply it -- when that happens, consumers revert to robotic language and simple phrases, and developers just give up."
To use Alexa Conversations, developers give Amazon a few sample phrases of their conversations as well as some APIs that implement the services they're trying to achieve. From those samples, Amazon's AI system tries to anticipate all the possible conversation paths the user might take. It reduces the amount of back end code developers have to create and the amount of training data they have to provide.