OpenAI has introduced a ‘study mode’ for ChatGPT to foster responsible use in education, countering rising concerns about AI misuse in schools and universities. The feature, designed to assist students with homework, exam preparation, and topic exploration, enables interactive, step-by-step guidance akin to classroom learning. OpenAI stated on its blog that the goal is to help learners “understand and analyze material” rather than depend on pre-packaged solutions.
For instance, when a user sought help with Bayes’ theorem, ChatGPT first queried their math proficiency and learning objectives before delivering an explanation. Ayna Devani, an OpenAI manager, emphasized the tool aims to let students “interact with ChatGPT to tangibly support the learning process.”
The rollout follows heightened academic alarm over AI-aided cheating. A recent Guardian investigation uncovered nearly 7,000 verified cases of university students using AI tools to cheat during the 2023–2024 academic year. Concurrently, U.S. data reveals over a third of college-aged adults use ChatGPT, with roughly a quarter of all messages to the bot involving learning, teaching, or assignments.
“We don’t believe these tools should be misused, and this is a step in that direction,” Devani added. OpenAI developed the feature with educators, scientists, and pedagogy experts but cautioned that “inconsistent behaviors and errors in conversations” may occur.
The move coincides with rival Anthropic launching a similar “learning mode” for its Claude chatbot, while Google unveiled new student-focused AI tools in search and its Notebooklm platform.