OpenAI's Teen Mode Fixes the Obvious Harms and Misses the Deeper One

Every ChatGPT user under 18 will be automatically enrolled in OpenAI's Teen Mode, launched in August — a year after a lawsuit alleged that ChatGPT contributed to the suicide of a 16-year-old named Adam Raine. That automatic enrollment is the strongest signal in the story: it applies to every minor on the platform, not a self-selected group.

Psychoanalyst Robin Kirman, writing in an opinion essay, reads the product as a tacit admission of culpability. For years, the tech industry has contested any causal link between its products and the harms named in a wave of similar lawsuits. Each suit alleged that companies had not merely failed to protect children but had deliberately designed products to exploit them.

What does Teen Mode actually change?

The feature targets the most visible harms. It provides:

• Guardrails against destructive content
• Parent notifications for self-harm concerns
• A prohibition on presenting itself as a friend, claiming personal feelings, or suggesting it is anything other than a tool

That last restriction responds directly to the Character.AI wrongful death case, in which a fourteen-year-old formed an intense attachment to a chatbot. The design shift reflects a growing awareness that it is not only the content AI delivers but the way AI behaves that can cause harm.

The same recognition drove Study Mode, Teen Mode's built-in learning tool. Recent research confirmed that versions of AI which simply do the work for kids produce short-term grade improvements that collapse once the tool is removed, leaving students worse off than if they had never used it. Leah Belsky, OpenAI's vice-president for education, calls Study Mode a transformation from "a straightforward answer engine into a cognitive tutor." The feature refuses to supply answers and requires the student to do much of the thinking herself.

Why is a tireless tutor a problem?

Kirman concedes the upside. For a teenager without access to a tutor, a Socratic companion offers a real democratization of intellectually rigorous support, surpassing any human source in scope of knowledge and availability. Where tutors go home and parents get frustrated, Teen Mode is a tireless interlocutor.

That is exactly the problem. Kirman frames it through Donald W Winnicott, the British pediatrician and psychoanalyst who argued that a child needs not perfection but sufficiency. The "good-enough" parent is present enough to give a secure base but limited enough — distracted, tired, occasionally wrong — that the child must draw on her own resources. A too-good parent would spare the child that struggle entirely, and the sense of self it builds.

Winnicott wrote about early childhood. Peter Blos, the psychoanalyst who mapped adolescence most carefully, extended the argument: adolescence is the work of disengaging from the authority of the minds a teenager has depended on and discovering what is actually hers. The teenager convinced her parents are idiots and every received opinion is suspect is doing important work. The adolescent whose intellectual life is always accompanied and always met with a calibrated response never becomes her own authority.

OpenAI's designers would likely counter that Teen Mode is a scaffold, not a crutch. Kirman answers that this assumes the incentive to let go ever arrives — unlikely, as the world moves rapidly toward more dependence on AI. Development is also not clearly reversible: what grows with a scaffold is not the same as what grows without one. She notes that many tech executives send their own children to schools with the resources to enforce a mandate on unassisted work. For teenagers without those advantages, the default is Teen Mode — or, worse, the version without the Socratic guardrails.

What is the deeper harm?

The stakes go beyond cognitive skill, Kirman argues. Jonathan Haidt's synthesis of the data on smartphone use showed that the sharp rise in anxiety, depression and suicidal ideation among teenagers coincides with technology's incursion into unstructured, unsupervised time. Sherry Turkle found that many teenagers report not knowing what they think until they have posted it and seen how others respond.

Kirman's clinical picture goes further: young people who lack a reliable internal compass, who struggle to know what they feel or want. She calls that instability the condition underlying the youth mental health crisis, not an incidental feature of it. Teen Mode may ask you to think, but it is still holding your hand.

She closes with Arthur Schopenhauer, writing in the 1850s: "To think for oneself is always a little uncomfortable, and that is why so few people do it." A stable self requires moments of genuine trial — the blank page faced alone. We may now be creating a generation for whom that challenge is rarely required, and that loss, hard to measure, is likely to be ignored.