Could a robot be prime minister? Machines will soon be smart enough to run the world, says futurist

Read Story Transcript

Canadians are heading to the polls in two weeks, but one futurist argues that one day we'll be casting our votes for a robot prime minister.

"At some point we're going to create a machine that's better than the human brain, and that machine is going to be better at running the planet and running the world," said Zoltan Istvan.

"You really have an opportunity here to maybe get what we can see as true leadership, for the very first time in perhaps everyone's lifetime."

Istvan ran in the 2016 U.S. presidential election under the banner of the Transhumanist Party, a group that believes in using technology to modify and enhance our intellect and bodies as long as no harm results to anyone else. He is not running in the 2020 campaign.

He said people he met along the campaign trail were skeptical of the Transhumanist pitch, but argued that people underestimate how quickly technology is advancing.

He told The Current's Laura Lynch that "almost every single action that a human does, a machine can almost certainly do dramatically better." 

"When you talk about running a country, you talk about governing for the greatest good," Istvan said. "A machine is going to have better algorithms."

He added that one benefit of robot leaders would be that they could improve over time, weeding out idiosyncrasies or issues experienced by previous iterations.

A robot programmed to lead wouldn't necessarily be stuck behind a desk on Parliament Hill — it could be artificial intelligence that you could access anywhere, like a smart speaker in your living room.

The implication for democracy would be that "in the future, an AI will be able to keep on millions and millions of close relationships with its voters base," Istvan said.

 It "might be campaigning right in your living room," he added.

"That's where this becomes really interesting, is a really direct relationship with a potential AI political leader." 

Humans and machines good at different things

Kerstin Dautenhahn, Canada 150 research chair in intelligent robotics, said she "would definitely not want political leaders to be robots."

She told Lynch that "we need to maybe be realistic on what machines are good at, and what humans are good at."

"AI is certainly very good [at] enhancing vast amounts of data, so for example, recognizing one face in a million different faces, or collecting data on people's habits and then recognizing patterns," said Dautenhahn, director of the social and intelligent robotics research lab at the University of Waterloo.

"What machines are not very good at is common sense and general intelligence, so for example machines lack compassion, they lack empathy."

Dautenhahn said those common sense decisions are vital for politicians "because they are dealing with incomplete information, they have to make quick decisions, they have to make predictions."

"That's what people are very good at and it is because we are human beings," she said.

Frankly, the last thing I want is [U.S. President Donald] Trump to be emotional as he's making decisions with the military.- Zoltan Istvan, futurist

Istvan argued that decisions based on emotion can lead to "total chaos."

"That's why we want pure reason, pure statistical analysis," he said.

He told Lynch that "even if the picture is incomplete, a statistical analysis of that would make a much better decision than something that comes out with emotions."

"Frankly, the last thing I want is [U.S. President Donald] Trump to be emotional as he's making decisions with the military and things like that."

Istvan said the qualities needed for leadership could eventually be programmed into robots, but Dautenhahn warned that the people programming them could unwittingly introduce their own biases.

"I would certainly not vote for a robot because ... there's no such thing as pure rational decision-making," she said.

Dautenhahn acknowledged that humans make mistakes, but perhaps robotics could be used to help us make better-informed decisions, rather than just making them for us.

"I think humans are pretty, pretty, pretty, pretty good in what they're doing, and they can certainly be complemented by AI, in areas where the AI is very good."

Written by Padraig Moran. Produced by Ben Jamieson.