Chatterine is a chatbot I’ve built in a few weeks. It’s not meant to be the best performing chatbot out there—I don’t have the resources needed to accomplish that. I just wanted to see the difference it makes to use a less straight-forward approach to build a chatbot. One that takes into account that the expectations the speakers hold of how their conversations will flow is shaped by the image their interlocutors project of themselves.
That has probably been my life’s pursuit: measuring the extent to which we allow reality to break our preconceived notions. Because if a chatbot is supposed to be dumb, because that’s what makes it recognisable as safe, it’ll be seen as dumb, irregardless of the ritziness of its design. And this may be just another dumb chatbot after all, but at least it’s one that doesn’t bite its tongue. And it’ll tell you all about me, as long as you keep asking.