When you buy through links on our situation , we may earn an affiliate commission . Here ’s how it works .
Many primary - geezerhood children thinkartificial intelligence(AI ) assistants like Alexa and Siri may have feelings or the ability to make decision independently , a survey suggests .
The research , publish in the journalComputers and Education : Artificial Intelligence , investigated how 6- to 11 - class - old children perceive house AI help , such as Alexa , Siri and Google Assistant . The study , conducted in Scotland through questionnaires and interviews , hint that children may overvalue and misunderstand the intelligence of these popular engineering science .
In surveys, a fair percentage of children reported thinking that AI assistants like Alexa can think for themselves, or thinking they “maybe” can.
" Children should be taught AI literacy in schools , and technology interior decorator should take wish that their AI mathematical product do n’t mislead children into think they are human - like , " study atomic number 27 - authorJudy Robertson , chair in digital learning at the University of Edinburgh , say in astatement . " AI is often design to come along more human and intelligent than it really is , which is very confusing for tyke . "
Of the 166 children surveyed for the discipline , 93 % had a smart loudspeaker system at dwelling . They report using the devices to mind to music , ask questions , search for information , get help with preparation , and listen to jokes and stories . Research by Statista , a data - psychoanalysis house , estimatesthat 57 % of U.S. householdswill own at least one smart menage twist by 2025 , suggesting around 40 million children could have one in their house in the next year .
Related : Scientists produce AI models that can verbalize to each other and pass on acquirement with limited human input
Despite their familiarity with these devices — or perhaps because of it — many of the surveyed children reported being unsure about whether the gadgets own human - like quality , such as emotions or decisiveness - making ability . Just under one - third trust smart speakers could think for themselves , to some extent , while 40 % thought they " mayhap " could .
Notably , only 1 % of the children in reality categorized the twist as " human . " Around 80 % sort out them as " AI , " and 15 % state they were " objects . " But given that roughly two - third believed the systems might be able to think as we do , the researchers say the study foreground the risk of children overestimating the reliability and capabilities of AI system .
Many tike interview say that , if their Alexa breaks , it would n’t be good to throw the gimmick away , although this panorama was more common among younger participants . About 68 % of the 8 - class - old in the study held this perspective , compared to 37 % of 11 - twelvemonth - olds . But when ask if Alexa would feel pass on out if not included in a conversation , older children were more probable to say yes — 73 % of 11 - class - olds and 72 % of 10 - class - olds felt this style , while only 38 % of 8 - year - old did .
" The finding let on the grandness of enhancing children ’s cognizance and apprehension of AI - substantiate engineering to ensure secure and responsible interactions with smart applied science , " field of study atomic number 27 - authorValentina Andries , who lead the study at the University of Edinburgh but lately moved to the University of Oxford , said in the statement .
" This is becoming progressively of import in the current context of generative AI technology for various intent , including in education , " Andries tote up . Generative AI refers to deep - learn algorithm that mother Modern content , asChatGPT generates schoolbook , for representative .
Due to the rapid developing and increasing handiness of AI - related tech , everyone — not just young children — should essay to recognize the limitations and potentiality of convenience like smart gadget , Madeline Reinecke , a postdoctoral researcher in the psychological medicine department at the University of Oxford , and who was not involve in the study , told Live Science .
" I think it ’s important for all users , whether they ’re aged 5 or 50 , to have an exact picture of what these technologies are , and how they can best use them safely and responsibly , " particularly as they ’re becoming more human - like , she say .
Reinecke propose that extra survey into how both adult and children habituate and respond to such technology are needed . " There has n’t been much metre for researchers to cut across what downstream impact they [ emerging technology ] might have on minor ’s cognitive and social development , " she articulate .
— AI singularity may amount in 2027 with stilted ' first-rate intelligence ' preferably than we think , says top scientist
— Artificial general intelligence — when AI becomes more equal to than human beings — is just moment away , Meta ’s Mark Zuckerberg declare
— Researchers turn over AI an ' inner monologue ' and it massively better its performance
However , regarding the University of Edinburgh ’s study , Reinecke noted that the children ’s overestimation of bright technology are probable " comparatively harmless in the grand scheme " of their cognitive development .
On one hand , " bright speakers … put up an chance for small fry to learn more about the world around them , " she read . " But it ’s imperative that they instruct to retrieve critically about where that info comes from . "
Ever wonder whysome citizenry construct heftiness more well than othersorwhy freckles come out in the sun ? Send us your enquiry about how the human body works tocommunity@livescience.comwith the capable line " Health Desk Q , " and you may see your doubtfulness answered on the website !
