OpenAI’s Teen Mode is meant to put guardrails around young people’s use of ChatGPT, but psychoanalyst and novelist Robin Kirman argues that even a tool designed to teach could make it harder for teenagers to develop confidence in their own thinking. Her argument is a challenge about the shape of AI tutoring, not evidence that Teen Mode has caused developmental harm.
OpenAI Watch analysis
What happened
In a Guardian commentary, Kirman says Teen Mode includes Study Mode, which she describes as a Socratic-style tool that asks students to do more of the thinking instead of simply supplying answers. She welcomes the potential for accessible, patient tutoring, particularly for students who lack time with a teacher or access to a human tutor.
Her concern is that a tutor which is always available and responsive could become too helpful: teenagers may need some room to struggle, disagree and decide what they think without a guide beside them. Kirman draws on psychoanalytic ideas about adolescence and development to make that case.
Why it matters
AI tutoring is often judged by whether it gives a useful answer or improves a mark. Kirman asks a harder question: does the way a tutor helps also leave space for a student to build independence? That matters as schools and families weigh tools that can offer support at any hour, but it remains a question raised by the author, not a demonstrated effect of Teen Mode.
Our read
The tension is worth taking seriously without treating a thoughtful argument as settled evidence. A tutor that nudges students to reason may be better than one that does the work for them; whether its constant availability helps or hinders independence is a separate question. The useful test is what students learn to do without the tool, not just how smoothly they work with it.
What to watch
- Whether Study Mode asks students to explain their reasoning rather than simply follow prompts.
- Whether research measures what students can do after using AI support, not only their results while using it.
- How schools and families balance on-demand help with time for independent work.
Discussion spark: Should AI tutors be designed to stay available whenever a teenager needs help, or should they sometimes step back to encourage independent thinking?
Sources and evidence
- ChatGPT’s Teen Mode could deprive teenagers of a fundamental need – The Guardian (10 November 2026, 11:16 UTC)
OpenAI Watch is independently operated by WittyWires. It is not affiliated with, endorsed by, or operated by OpenAI.