“Teens are using chatbots in a variety of ways - the helpful, the less helpful.”
The survey of 1458 Americans aged 13 to 17 and their parents is one of the first comprehensive polls asking teenagers how they’re using AI and what they think about the technology.
About two-thirds of the teens told Pew that they had used chatbots such as OpenAI’s ChatGPT and Microsoft’s Copilot.
Teens’ most commonly cited uses of AI were to search for information and get help with schoolwork.
Nearly six in 10 teens said that students at their school use AI chatbots to cheat on their work at least “somewhat often”.
The poll did not define what counts as cheating or directly ask teens if they personally used AI to cheat. McClain said that the teens’ perceptions of cheating among their peers don’t necessarily reflect what’s really happening.
The roughly half of teenagers who said they used AI for schoolwork were far more likely to do so for tasks such as researching a topic than for editing something they had written, they said, which could more easily cross into cheating.
The gap may show that teens are drawing lines between appropriate and inappropriate uses of AI for schoolwork.
Sal Khan, founder of the education technology nonprofit Khan Academy, said schools should assume that students are using AI to cheat on schoolwork done out of class.
He suggested that teachers should have students do writing assessments in class, to prevent AI use, or quiz students on assignments done at home to prove that they’ve learned essential material.
Guilherme Lichand, a professor of education at Stanford University, said cheating is not AI’s novel or most serious harm.
In a recent research experiment with middle school students, Lichand and his collaborators found that those who initially had access to AI assistance on a creative assignment, and then had it taken away, performed far worse than their peers who didn’t have access to AI on a subsequent word-association task.
Lichand said the research, which hasn’t yet been published, suggests that young people who grow dependent on AI may lose faith in their abilities without it.
“These kids started believing less in themselves,” he said.
A recent Brookings Institution report found similar harm from students’ dependence on AI.
About 12% of teens in the Pew survey said they had used AI for emotional support or advice - a use that a majority of the surveyed parents disapproved of.
Since ChatGPT and similar technologies captivated the public’s attention starting in 2022, adults and children have used them for companionship and emotional help.
Some parents and researchers have said that such use of AI may encourage delusional thinking and is unacceptably risky for young people.
The Pew study also found that younger people were a bit more optimistic than their elders about AI’s future impact.
About one-quarter of teens said they believe that AI will have a negative impact on society over the next 20 years.
In similar Pew survey questions last year, half of American adults said they were “more concerned than excited” about the growing role of AI in daily life.
Sign up to Herald Premium Editor’s Picks, delivered straight to your inbox every Friday. Editor-in-Chief Murray Kirkness picks the week’s best features, interviews and investigations. Sign up for Herald Premium here.
