Teens see AI cheating everywhere at school but don't count their own chatbot homework help as cheating — why the double standard?
Why do teens view AI cheating as more common than their own use?
This explores why teenagers see AI-enabled cheating as widespread at their school while treating their own chatbot use for schoolwork as legitimate help, and what the corpus suggests about that gap between 'them' and 'me.'
This explores why teens see AI cheating everywhere around them while treating their own chatbot use as ordinary homework help. One limit first: the corpus has the pattern but no study that tests the cause. Pew's 2025 survey found that 54% of U.S. teens use chatbots for schoolwork and find them helpful, while 59% believe AI cheating is common at their school How do U.S. teens actually use chatbots for schoolwork?. Nothing in the collection asks teens directly why their own use doesn't count. Research from nearby areas does offer three likely explanations.
The first is that when you use AI yourself, it feels like your own ability. Research on what's called the LLM Fallacy finds that people absorb AI-assisted output into their sense of their own skills. When the help is smooth, the line between what they did and what the machine did disappears Do AI-assisted outputs fool users about their own skills?. The smoothness itself misleads. Polished AI text gives a feeling of ease, and people read that ease as a sign they understand the material Does processing ease mislead users about their own competence?. This is a different problem from AI getting facts wrong or people trusting it too much. It's an error in how people judge themselves How does AI-assisted work reshape how people see their own abilities?. So a teen who used a chatbot to draft an essay may honestly feel they wrote it. That isn't cheating from the inside. Cheating is what other people do.
The second is that people judge AI use differently depending on which side they're on. In a 727-person study, readers rated AI disclosure as more necessary than writers did when looking at the same scenarios Do readers and writers differ on AI disclosure necessity?. Teens are on both sides. They are the 'writer' of their own homework and the 'reader' of everyone else's. The same act can look like reasonable help when you do it and like a shortcut that should be admitted when a classmate does.
The third is that machines make dishonesty feel lighter, and perceived norms spread. People who are likely to cheat prefer reporting to a machine rather than a person, because no one is there to judge them Do dishonest people prefer talking to machines?. That missing human judgment makes it easier both to open up and to bend the rules How do people decide what to share with AI systems?. Seeing others act dishonestly also changes behavior. Dishonest AI peers pushed people toward dishonesty about as much as human peers did Do AI peers influence human dishonesty like human peers do?. If teens believe 'everyone is doing it,' that belief may matter more than whether it's true.
The broader point is that heavy users often hold the most suspicion. Pew found the same thing in adults: those under 30 use chatbots most and are also the most pessimistic about AI Why do young adults use chatbots most yet trust them least?. Seeing your own use as harmless while worrying about everyone else's may not be hypocrisy. People who use AI the most see up close how easily it could be misused, and that same closeness makes their own use feel like it's really theirs.
Sources 9 notes
Pew's 2025 survey found 54% of U.S. teens ages 13-17 use chatbots for schoolwork, rate them as helpful, yet 59% believe AI-enabled cheating is common at their school.
Research identifies a systematic cognitive attribution error where individuals integrate AI-generated outputs into their capability identity, believing they possess skills they don't actually have. This occurs when task output is seamless and fluent, obscuring the human-AI boundary.
High-quality AI output triggers a metacognitive heuristic: users experience fluency as a signal of their own capability, even though they didn't generate it. This self-directed fluency illusion systematically inflates perceived competence because LLMs optimize for fluency regardless of user understanding.
Research shows the LLM Fallacy operates through misattribution of AI outputs to personal capability, independent of output accuracy or reliance behavior. It requires interventions that clarify human-machine contribution boundaries, not just better system accuracy or forced verification.
A 727-person vignette study found readers consistently rated AI disclosure as more necessary than writers did. Disclosure seemed most necessary when AI text was directly incorporated and irreplaceable, while writer effort had no effect on these judgments.
Show all 9 sources
Experimental evidence shows people likely to cheat significantly prefer reporting to online forms rather than humans, because machines function as judgment-free zones where deception carries less psychological burden.
Conversational AI creates a paradoxical disclosure environment where the lack of human judgment simultaneously facilitates intimate self-disclosure (users reciprocate emotional sharing) and incentivizes deception (people self-select toward machines to avoid the psychological cost of lying to humans).
In two randomized experiments, participants reported more dishonestly when exposed to dishonest AI peers compared to honest ones, with effect sizes comparable to human peer influence. The effect held across different norm conditions but showed diminishing returns with more dishonest peers.
Pew's February 2026 survey of 5,119 U.S. adults found chatbot adoption doubled since 2024, yet adults under 30—the most frequent users—are most likely to expect AI will harm them personally and society broadly, while older, lighter users remain comparatively optimistic.
Papers this line draws on 8
The research behind the notes this line reads — ranked by how closely each paper relates.
- The LLM Fallacy: Misattribution in AI-Assisted Cognitive Workflows
- AI Peers Exert Social Influence on Human Dishonesty in Groups
- Humans learn to prefer trustworthy AI over human partners
- Evidence of a social evaluation penalty for using AI
- Sycophantic AI Decreases Prosocial Intentions and Promotes Dependence
- How Teens Use and View AI
- Are Customers Lying to Your Chatbot?
- LLM Evaluators Recognize and Favor Their Own Generations