To start the year we take up a topic suggestion from our apprentice Arturs and examine whether chatbots could replace conventional search engines in future.
Most of us use them daily: search engines. Whether Google, Bing, Yahoo or one of countless others, they have become impossible to imagine our daily lives without. Your son wants to know which dinosaur was the fastest? You need a recipe for the next family gathering? And what will the weather be like at your holiday destination? Bill Gates's well-known phrase “information at your fingertips” has become a comfortable reality for all of us. We click through links and websites and find the information most useful to us in no time.
But what about our private data or our work material? When it comes to the details of your own mobile contract or a company's internal policies, search engines are of limited help.
Chatbots can take the way we get information to the next level. There are already ways of connecting search engines and language models. The answer to a question is then no longer simply a list of links but a formulated answer in natural language with a source reference. And there is more. Via API, a range of further systems can be connected to chatbots, giving access to the information they hold. For companies that means staff can access all relevant information or files quickly and easily – no more long searches across various filing and information systems.
We are sure this progress will find its way into private life in the near future too. The chatbot may then not only announce the weather at your holiday destination but also know the details of the booking and, if needed, the terms of your travel health insurance. Anyone who has tried to switch to the best possible mobile contract knows that the comparison alone can take hours out of your life. How pleasant it would be if the chatbot on your smartphone not only found the best provider but also knew your notice period and drafted the cancellation letter for you.
A word of caution, though. Not all the obstacles have been cleared. All currently available language models are prone to hallucination – above all when the information sought is not available. This phenomenon is currently occupying researchers and developers worldwide. At Splitbot we increase the reliability and verifiability of the answers delivered by citing the sources used with every answer.
Thanks to the rapid development of chatbots and language models, though, the end of hallucinations is within reach, and it will be fascinating to see which areas of life chatbots move into next.
