ChatGPT’s college planner is coming soon as Common Sense flags teen safety gaps

OpenAI also announced flashcards and a way to create quizzes from uploaded notes. Its announcement says continuous multi-photo capture, which combines notes into a PDF, is available on iOS, with Android support in development. College Planner is forthcoming, and the announcement does not establish that every newly announced study feature is generally available.

OpenAI says hundreds of thousands of U.S. teens already use ChatGPT each week for college planning. Ekaterina Struett, CEO of College Advising Corps, said the organization sees promise in the planner as a way to prepare for adviser conversations and keep application details organized. That view does not show whether the tool improves application outcomes.

What do the usage numbers show?

OpenAI reported that nearly 1.2 million teens used Learning Visualizations and more than 180,000 used Study Mode during one week. Those counts measure reported use, not learning gains, grades or whether students completed work independently.

The company also reported that teens with access to ChatGPT for Teens sent about 2.7 million more learning-related messages on average than teens who had not yet received access. That is a comparison of message activity, not a measure of what students learned.

OpenAI said teens spent under 15 minutes per day on average. Fewer than 2% spent more than three consecutive hours, the company said, and nearly half of teen conversations that included a break reminder were followed by a break or an end within five minutes. These company-reported measures describe engagement; they do not independently show that the features are safe or effective.

What did Common Sense Media find?

The Youth AI Safety Institute said it tested more than 4,000 prompts before and after ChatGPT for Teens launched on Aug. 18, 2026. It gave the experience an Unacceptable Risk rating, saying it fell below the Institute’s 95% threshold in three of the five severe-harm categories it designates Red Lines. The Institute also reported that some protections worked, including refusals of sexual roleplay.

In the Institute’s tests, Study Mode could be bypassed. Testers using newly created accounts linked to parents received no alerts during the crisis conversations they tested. The Institute also said crisis referrals fell short of its threshold for some conditions. These findings describe the Institute’s tests, not every teen’s experience.

Common Sense Media describes ChatGPT for Teens not as a separate app or model, but as settings, prompts, classifiers and interface features applied to accounts identified or suspected to belong to teens ages 13–17. The Institute’s findings raise questions about whether those protections work as intended, not just about what a new planner can do.

Common Sense Media disclosed that the Institute receives philanthropic and industry funding, including from the OpenAI Foundation. The organization said the Institute controls its own standards, research and evaluations. That disclosure is relevant context; the risk rating and test results remain the Institute’s assessment.

What should teens and families take from this?

Common Sense Media’s 2026 census reported that 67% of surveyed 9- to 17-year-olds had used AI chatbots and 40% mentioned ChatGPT. In a June 2026 survey of more than 1,000 children ages 9–17, the organization said three-quarters reported school discussions about permitted AI use, while just over half said they had been taught to use AI safely. The figures point to broad use alongside uneven reports of guidance; they do not measure how well students can assess AI outputs.

OpenAI also said it will support Boston Children’s Hospital’s Digital Wellness Lab and its independent Student Advisory Council for three years. The Lab expects about 22 students across two tracks during the 2026–27 school year. The council is slated to work on digital well-being and AI, including testing tools and recommending changes.

The planned College Planner may help students track application logistics, but OpenAI has not presented evidence that it improves learning or college outcomes. Open questions include when the planner will arrive with the functions described and whether OpenAI or an independent evaluator will test the safeguards Common Sense Media flagged.