Kids aged 10-18 report mixed but nuanced views on AI use and risks in daily life

The study captures a generational divide in AI engagement that quantitative data obscures. While surveys show high adoption rates among teens, interviews reveal that many use AI reluctantly or sporadically, often because it’s embedded in tools like search engines rather than sought out. This passive adoption suggests engineers may overestimate enthusiasm for standalone AI products among younger users. The indifference reported by some kids, expressed as dismissive language or outright refusal to discuss AI, hints at a potential backlash if tools feel imposed rather than useful.

Kids articulate specific technical and ethical failures that engineers should address. Teens cite examples like AI-generated code being too sloppy for game development or chatbots enabling account takeovers, showing they notice flaws adults might overlook. Their concerns extend beyond functionality: some reject AI due to environmental costs or fear it will replace creative problem-solving. These critiques align with broader debates about AI’s energy use and societal impact, but the youth perspective adds urgency to designing for sustainability and human agency.

The findings highlight a mismatch between how kids use AI and how adults assume they do. While adults worry about cheating or emotional dependence, kids report using AI for mundane tasks like homework help or programming experiments. They also express skepticism about AI’s current capabilities, comparing it to early, crude versions of transformative technologies. This suggests engineers should focus on incremental, practical improvements rather than revolutionary applications. The study also reveals that kids are more concerned about AI’s societal impact than personal job displacement, a nuance often lost in adult-centric discussions.

Schools’ responses to AI, like shifting to in-class writing, create unintended consequences for students. Teens report that anti-cheating measures, such as time-limited essays, hinder their ability to produce quality work. This feedback loop, where AI adoption leads to restrictive policies that degrade learning experiences, is a cautionary tale for engineers. Tools designed to assist students may end up forcing educators to adopt counterproductive measures, undermining the original goal of supporting learning. The study underscores the need for AI tools that align with pedagogical goals rather than disrupt them.