Unfortunately, that’s a double-edged sword. While these systems are flexible and able to learn a great deal on their own — you can’t just insert an extra fact, or explicitly handle an edge case.
We don’t know how to build conversational software yet

:-) that’s the old “stability-plasticty dilemma” that we faced with neural nets learning… that’s why I’m a bit perplex when I read about a “quick & dirty” proposal of deep learning as a chatbot dev enabler now, in 2016…

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.