Chatbots: Past, Present and Furture

instant web app as chatbot
"Who controls the past controls the future: who controls the present controls the past." - George Orwell

Just 2 years back the techno gurus where announcing chatbots as next big thing, like this article in medium: 9 Reasons Why Chatbots are the Next Big Thing"

My first experience with chatbots was with Alice chatbot several years back in 2006. I experimented and tried to use it for some useful things, but it was just a bit short of being useful. It was good entertainment though during those days. I was able to hook it up with a multi-player cricket game chatbox.

Things have changed a lot since then, there are a new variety of chatbots, some are just web forms masquerading as chatbots like the one banks and insurance companies are running. Yes, they are bit smarter than filling up forms and better at keeping the user engaged. These kind of chatbots are useful and found on many websites now. The main USP is that instead of digging into the complex maze of webpages, they provide a single point of initiating communication. They are good at hiding complexities and in the process, let the customer interact smoothly with the services a website provides. I find the customer service chatbot on pretty helpful. Amazon customer service chatbot gets the required details from you helps you sort thru tons of support materials and seamlessly passing control to a real person once it has run out of options.

Though chatbots have been finding a good business reason to thrive there is one area where that have lagged severely. If a user wants to hold a conversation like he will do with a real customer service agent then he is in for a disappointment. None of the chatbots that masquerade as a real person can go beyond few lines before getting discovered. So it is counterproductive to masquerade. It is better that a person knows that he is talking to a bot and be prepared to answer or question inline with the bot's preferences.

There is space for both Apps and Bots. They should be complementing each other. A hybrid approach ie, incorporating chat into a broader strategy that encompasses more traditional elements, will be more effective. A multi-model model where you combine chatting with email and text and with data on the website is going to be the most useful.

My experiments with Augmented reality started in 2010. It was an upcoming and very promising technology. Still Augmented reality till date is mostly good for games. It is not simple to productise augmeneted reality for enterprises. Recently it is gaining traction in creating visual reality conference rooms, where people can actually see models and discuss about them. Chatbots in some way are like augmented reality. Both these technololgies are exciting and gaining ground, but they still have to take some leaps, to be accepted by broader world.