2017: Will This Year Witness Bot-To-Bot Communication?

In the dynamic world of technology, bot tools and artificial intelligence have already made their mark. Everyone wants to learn how to make a bot that can simulate natural human conversation. Chatbots are now implemented on most popular messaging channels, such as emails, push notifications and in-app messaging. They’re also implemented on websites in the form of web widgets.

But something that’s generating more buzz than human-bot conversation is bot to bot conversation. January 2017 saw more that 3.5 million viewers watch two modified Google Home speakers, Vladimir and Estragon argue about the meaning of life, love, religion, and even their own existence.

Silicon Valley, which is home to several global technology companies, including Apple, Facebook and Google, is experiencing a new paradigm shift in how chatbots function. A lot of roboticists are working to build machines that can actually carry out conversations with each other, along with humans. According to them, it’s a necessity for software bots to build their own language, unlike the language that Vladimir and Estragon communicated in.

There may be a question in your mind - why do bots need to interact with each other in the first place? To understand this, we need to understand the complex universe of technology. Bots are set to complete a fixed set of tasks, such as moving to a landmark, and in order to do that they must collaborate with each other to complete the required tasks.

So how does this happen? The underlying principle is reinforcement learning, which bots work on. They function on trial and error, and make a note of what works and what doesn’t as they try to complete the task assigned to them. They assign characters to concepts, locations, objects, actions, and to each other. Any action that works for them, needs to be done again whenever they have to do the same task. This is how they build their language and converse with each other, helping them get to their destination faster.

In the end, these methods will give the software, which bots rely on, a better understanding of languages, and help bots figure out why language exists. For humans, the need for communication stemmed from necessity to do a lot of things, and roboticists want to explore the same dynamic for these computer programs. If not actual language, even actions and gestures that bots use to guide each other is a significant step in this direction of collaboration.

Achieving these milestones will take time, and will happen by using a lot of different techniques. 2017 is touted to be the year where researchers try a combination of techniques, and devise a method by which bots can learn to chat in their own language.

Chatbots can be used to interact with your customer in customer service process easily.