There are two main kinds of chatbots: rule-based and AI chatbots.
Rule-Based Chatbots use simple predefined sets of questions and answers to take a person along a straightforward path. They have specific tasks to complete and are not flexible. These are sometimes called ‘Click Bots’ as they work with basic button clicks and rigid replies.
AI Chatbots use conversational AI technology to have life-like interactions with humans. They use artificial intelligence systems to do this - Natural Language Understanding to read the intent of what a person is saying and Machine Learning to continually grow in their understanding of language and how to best respond.
There are three layers of AI automation at work in AI chatbots:
AI chatbots are particularly powerful in customer service settings and as virtual digital assistants, like Alexa. In customer service, the AI bots are taught to understand what a person is asking or wants to do, and then responds accordingly, without the need for a live agent to deal with the query. For more detail, read about the different chatbot examples.
A blended approach is often the best way to use AI chatbots. In this case, a human agent is available to deal with the more intricate issues that the bot cannot handle but leaving the bulk of queries to the chatbot to efficiently take care of for many people at the same time.