And that is not possible as for now.
Vladislav Arbatov
1

You have to define pretty much everything for the machine to have a meaningful conversation if it’s a retrieval-based model. And in case you want to make a generative model (when bot talks like a human with no phrases predefined)— it’s completely impossible with those “simple” publically available tools.

But. There’s a way to build a linear if/then story, if it’s your goal, with FlowXO service. No AI involved, plain scripting. I guess that could be what you’re looking for.

One clap, two clap, three clap, forty?

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