Speaker contrasts traditional rule-based chatbots (described as an "upside down tree" of predefined paths) with generative-AI service agents. The speaker argues rule-based systems force designers to enumerate every possible path, produce static, frustrating experiences, and that generative AI can deliver dynamic, human-like conversations that resolve user problems faster.
Before what you would have to do if you wanted to have a service agent you'd have to build a big tree. It looks like an upside down tree and that's all the different paths that it could go on and you would have to then pretty much create every possible scenario that it could be and what the end user is going to get is whatever you predefined you're building out the tree that they're going through. It's not dynamic and really what we want to be able to deliver is a truly. It feels like you're just chatting with a human because people aren't frustrated with chatbots. Everyone says they hate chatbots. It's a bad experience. What they hate is not getting the answer they want fast and not resolving their problems. So we're not against the idea of chatbots. I think we're against the idea of giving a bad experience and that's I think where we're going to see generative AI is really able to deliver.
