​Open Universities stays smart by keeping chatbots dumb

Disguising chatbots as humans is not a ticket to success for Open Universities Australia.

chatbots.png

Head to the Open Universities Australia site and you may encounter a chatbot. But rather than being the sort of annoying site popup that tries to convince users that a human is talking to them, StudyBot is not that at all.

"From day one we've been very deliberate about the fact that this is not a human, but we have a bit of fun with it -- if it doesn't understand something it'll say 'hang on, I don't get you, let me go get a human for you'," head of digital at Open Universities Australia Lyndon Summers said on Wednesday.

"We don't try and disguise it, we've actually gone out of our way not to even give it a personality, as such, that could be identified as an individual as opposed to an automated assistant."

Open Universities Australia uses StudyBot to present information to students, conduct simple triaging of problems, perform password resets, and pass basic authentication to call centre staff -- who were used to create the scripts it would say.

The end result is that the bot only scores a few points behind call centre agents, Summers said, which is testament to its simplicity and the efforts Open Universities Australia has made to ensure users understand it is a chatbot.

"If you set the right expectation then its okay, but if you dress it up .... and you are led down this path of thinking you are talking to a person, and it's not, that's only going to lead to disaster," he said.

"It's funny how often you'll see please and thank you in a bot conversation."

Also see: Open Universities turns to Bluewolf to redefine student experience

Summers added that Open Universities Australia would only keep conversations it deemed as high-value to humans to ensure "good interactions and good conversations".

Also making use of chatbots for triaging is health care multinational Bupa, whose general manager of digital growth Nick Blatt said that he had yet to see a chatbot do a better job than a human. Instead, the chatbots have been used to complement human operators.

"We are letting humans be more emotive, have complicated conversations, and the more simple things like 'Tell me how to log into my account' those kind of things, that's where we add value," he said.

"Something like Watson, even the machine learning in that environment, I just don't think it works really well, just yet, in my experience with it. It's very hard -- I still think you could get a bunch of people that would do a better job."

AI could be more harmful than social media

Also speaking alongside Summers and Blatt was LivePerson CEO and founder Rob LoCascio, who issued a warning that artificial intelligence has the potential to inflict more damage than social media on society.

LoCascio warned of the west coast technologist mindset of believing it could help the world.

"I think it is a very egotistical, self-centred perspective on how technology can help the world," he said. "I think government has got to get involved now and start helping shape what all this should be."

LoCascio said social media has proven it does not create a better community.

"Facebook, if I was the CEO, I'd shut it down. He's a billionaire ... it's not about money, it's something else. Shut it down, fix it, and then put it back up," he said.

"It's not good enough to let it run when we know it could be harming our community -- you're not making anything good.

"Shut it down tomorrow. Be a hero. You will win a Nobel Prize."

Related Coverage