By 1966, we had ELIZA — a computer program that could mimic the responses of a psychiatrist and, for short spurts, carry on convincingly human conversations.Those conversations were so convincing, in fact, that at least one person who worked at the MIT lab where ELIZA was being built would request to have time alone with it so they could speak in private. But it’s important to realize that, today, these human-substitute chatbots represent just one extreme of the chatbot spectrum.But in reality, today’s chatbots fall along a spectrum.
Ultimately, where a chatbot falls along the spectrum should depend on its particular purpose.
For example, it makes sense that the weather forecasting chatbot Poncho for Facebook Messenger is all the way over on the no-human/chatbot-only end of the spectrum.
So we landed somewhere in the middle of the spectrum.
Driftbot will engage with you to a point, but if the conversation gets off-track, the chatbot defaults to connecting you to a human.
For example, if you go to and start a live chat, Driftbot will ask you: “” Driftbot won’t try to figure out what those questions are — it will simply connect you to someone.
In the future, we might move a bit further toward the chatbot-only end of the spectrum and start using to Driftbot to ask people what their questions and issues are before connecting them with a human (provided that ends up helping people get answers more quickly).
It also felt impersonal, because it was context-irrelevant — it would say the same thing to everyone.
Our solution was to move a little bit closer to the chatbot-only end of the spectrum.
Drift product manager Matt Bilotti explained it to me like this: All the way on the right you have totally programmatic chatbots, where you give them specific prompts, or ask them specific questions, and they give you specific responses. It’s like when you call a customer support line and get that pre-recorded voice that asks, “What is the nature of your call, this, this, or this?
” and then you respond and get another set of questions, and another, and another.
While ELIZA couldn’t beat the Turing test, flash forward to 2014 and we have a stronger contender: At the Turing Test 2014 Competition at the Royal Society in London, the Eugene Goostman chatbot was able to convince 33% of judges that it was actually a 13-year-old Ukrainian boy. At Drift, we don’t want our chatbot to trick you into thinking that it’s a human.