Students and teachers suspect each other of using AI, Harvard research center finds

The Center for Digital Thriving describes a "two-way suspicion": students hide their use of AI, or even deliberately botch their work for fear of being wrongly accused.

The Center for Digital Thriving, attached to the Harvard Graduate School of Education's Project Zero, publishes work on how the generic AI transforms the relationship of trust into class. The team relies on a spring 2025 survey of more than 1,000 teachers and school heads of U.S. public schools, supplemented by interviews with students between April and June 2026.

A "double sense suspicion"

Researchers describe a climate in which everyone suspects that they are using AI, whether founded or not. Students say they feel compelled to hide their use of tools; Some go so far as to make work voluntarily less successful so as not to be falsely accused of cheating. For principal researcher Beck Tench, teachers are also losing visibility on learning: the questions that students once asked them are now going to chatbots.

The debate is already reflected in school policies: the public schools in New York introduced a moratorium on general AI for pupils from kindergarten to fourth (K-8).