We will love our robots. That much is hardwired into the human brain, according to Kate Darling, a research specialist at the MIT Media Lab.
From the heartfelt outpouring of sorrow when the friendly Canadian hitchhiking robot, Hitchbot, was vandalized in Philadelphia on its cross-US trip, to soldiers holding funerals for fallen robots that keep them alive, our attachment to mechanical creations runs deep, Darling said in a talk at the University of Waterloo this week.
So, as companies unveil robots that are deliberately designed to be personable, like the adorable social and customer service robot Pepper, and the singing, dancing, smiling LG Hub home assistant robot that made appearances at CES 2017 in Las Vegas, she wonders: How will these machines manipulate our emotions and change us?
It turns out that our affection for robots is not just a byproduct of watching loveable popular culture robot characters like R2-D2 in Star Wars.
"Our brains tend to be hardwired to project intent on any movement that happens in our physical space and that seems autonomous to us," Darling said. People are aware the machine is not alive. Yet they respond to the cues these lifelike machines give them, as if they were alive.
People have been known to give names even to simple robots that are not much more than sticks with wheels. But now, robots are being deliberately designed with movements, sounds and facial expressions that people associate with states of mind, she added.
One of the more sophisticated, lifelike robots is Sophia, developed by Hanson Robotics.
Darling says we are in era of human-robot interactions where the robots can influence people's behavior, for good or bad. This leads to all kinds of implications as more robots are deployed in homes, workplaces, schools, hospitals, nursing homes and other places.
Read More: This Silicon Valley Billionaire Wants to Give Us All Robot Bodies
One can imagine, for example, a home assistant interactive robot that can get people to reveal personal details they might not willingly enter into a database, Darling said.
People have even created panhandling robots. They were built for research or as fundraising tools, but the fact that people respond to cute robots and give them change, shows the technology could potentially be used to manipulate others.
Darling described animal-like robots like PARO from Japan, which looks a cuddly baby seal and responds with sounds, facial expressions and motions when touched. "It has been effectively used as an alternative to medication for calming distressed patients," she said.
She also cited the possible use of robots in behavior modification therapies.
The flip side is that robots could also be used to desensitize people to violence. "If people are taught to become violent toward lifelike robots, do they become desensitized to violence in other contexts?"
There many unanswered questions, Darling said. One of the most controversial issues is whether lifelike child sex robots might prevent pedophiles from offending, or should be banned because they might encourage it. "We don't really know," she said.
While prominent people such as Stephen Hawking, Elon Musk and Bill Gates worry that artificial intelligence could doom humanity, Darling says the legal and social implications of human-robot interactions will be more important in the immediate future.
"The problem might not be that if we teach a robot to kick a ball, the robot will come back and kick us. The problem might be that it does something to us if we kick the robot."
Get six of our favorite Motherboard stories every day by signing up for our newsletter.