A chatbot is a computer program designed to simulate conversation with human users via a messaging interface.
Thanks to the rise of artificial intelligence, they get a real understanding of what a person is saying and then generating the right response. However, chatbots can contextualize, remember and learn from conversation data to execute relevant future actions.
All of that is a new and efficient model for businesses to interact with their customers maximizing sales and support. Since users interact with the chatbot with a conversation, it builds an engagement that is more effective than any picture or video ad. Users can understand the brands better and marketers achieve the essential aim of advertising.
Chatbots can offer a wide range of services, from ordering flowers, booking appointments, movies, searching the web and much more, all in one place. Some real examples are Transferwise Bot which allows to send money via Messanger or Swelly that helps you to makes decisions with your friends who can vote for cool stuff.
Source: Transferwise Bot and Jessica Vomiero
The reality is, because of the lack of advancement thus far in NLP and ML technologies, chatbots are currently only able to handle basic interactions. Their communication with clients can seem canned. Early adopters claim chatbots have the same variety of conversational skills as automated call menus. Users can ask very general questions to a chatbot to get a very standard and short response.
Often it is unclear how to get started or they generate frustrating interactions since there is no menu of commands. Users does not have full command over their actions and requirements, you are at the mercy of a programming machine.
What happen when a user makes an error and the chatbot does not understand that you want to go back?
Source: Ted Carmichael
Even if Facebook Messenger developers have created 100,000 bots for Messenger in the first year of the Messenger platform, chatbots may not have had the impact that we eagerly anticipated just yet, but this doesn’t mean that they won’t deliver on their promise.
It seems Facebook absolutely overvalued what bots could do, in part because of all the excitement by the idea of artificial intelligent assistants who could do tasks and understand everything a user says.
Different chatbots were just shut down as well, on the 1st of August 2017, Facebook abandoned an experiment after two artificially intelligent programs appeared to be chatting to each other in a strange language only they understood.
Today only AI-powered virtual assistant as Alexa, Siri, Google Assistant, Cortana can be considered mainstream valuable products and they are apparently more on the way in the near future. At the same time, apps and web continue to reign over teach machines programming languages, based on conversation via text, because they offer wider contents and expanding your horizon.
The technology is in its early stage and only the future will say if it will be worth. Developers have to go further and implement newly advanced features which can create a closer human interface before a new technology is coming out.