Agentforce

Chapter 17How “Human” Should Your AI Agent Be?

That's an easy one.

It was answered nicely in the headline of an article posted by a seasoned product marketer a couple of months after the Agentforce launch: “How ‘Human’ Should Your AI Agent Be? In Short, Not Very.”¹

There is some hoo-ha in the market about where to draw the lines among the digital and human components of the orchestra. Because AI agents can now use some very natural language, it's tempting to see them as faux humans – a troop of Mini-Mes.

But at this stage in their (and our) evolution, this assumption is not only incorrect but misleading. The so-called Turing test of humanness in robot-computer communication is only useful as a commonsense trial of AI; it was never intended as a design principle.

In the early days of chatbots, some builders assumed they would simply impersonate people until they could not, at which point a real person would pick up the thread. This transition need not be noticed by the customer. But there is a phenomenon called the “uncanny valley” – that realm lurking between our real world and a facsimile that's almost right. This very quality of almost-ness disturbs us to our core processors. That is why Pixar films don't try for absolute photorealism.²

Very quickly, assumptions were changed. We humans are good at detecting nonhumans; by definition, we are experts in humanity. So it became standard practice for chatbots to identify themselves as “AI assistants” or the equivalent, much as ...