A latest examine carried out by Duke developmental psychologists explored how kids understand the intelligence and feelings of AI units, particularly evaluating the sensible speaker Alexa to the autonomous vacuum Roomba. Researchers discovered that kids aged 4 to eleven tended to view Alexa as having extra human-like ideas and feelings in comparison with Roomba.
The findings of the examine had been revealed on-line on April 10 within the journal Developmental Psychology.
Lead creator Teresa Flanagan was partly impressed by Hollywood portrayals of human-robot interactions, akin to these seen in HBO’s “Westworld.” The examine concerned 127 kids aged 4 to eleven, who watched a 20-second clip of every know-how after which answered questions concerning the units.
“In Westworld and the film Ex Machina, we see how adults may work together with robots in these very merciless and horrible methods,” stated Flanagan. “However how would children work together with them?”
Treating AI Units with Respect
Regardless of the variations in perceived intelligence between Alexa and Roomba, kids throughout all age teams agreed that it was mistaken to hit or yell on the machines. Nonetheless, as kids grew older, they reported that it was barely extra acceptable to assault know-how.
“4- and five-year-olds appear to suppose you don’t have the liberty to make an ethical violation, like attacking somebody,” Flanagan stated. “However as they become old, they appear to suppose it’s not nice, however you do have the liberty to do it.”
The examine revealed that kids usually believed that Alexa and Roomba didn’t have the power to really feel bodily sensations like people do. They attributed psychological and emotional capabilities to Alexa, akin to having the ability to suppose or get upset, whereas they didn’t suppose the identical of Roomba.
“Even with out a physique, younger kids suppose the Alexa has feelings and a thoughts,” Flanagan stated. “And it’s not that they suppose each know-how has feelings and minds — they don’t suppose the Roomba does — so it’s one thing particular concerning the Alexa’s capability to speak verbally.”
Flanagan and her graduate advisor Tamar Kushnir, a Duke Institute for Mind Sciences school member, are at the moment making an attempt to grasp why kids suppose it’s mistaken to assault dwelling know-how.
Implications and Moral Questions
The examine’s findings present insights into the evolving relationship between kids and know-how, elevating necessary moral questions relating to the therapy of AI units and machines. For instance, ought to mother and father mannequin good conduct for his or her kids by thanking AI units like Siri or ChatGPT for his or her assist?
The analysis additionally highlights the necessity to discover whether or not kids imagine that treating AI units poorly is morally mistaken, or just because it’d harm somebody’s property.
“It’s attention-grabbing with these applied sciences as a result of there’s one other side: it’s a chunk of property,” Flanagan stated. “Do children suppose you shouldn’t hit these items as a result of it’s morally mistaken, or as a result of it’s someone’s property and it’d break?”
