Product design question

Design parental controls for teen ChatGPT accounts. What should parents be able to do, what visibility should teens retain, how would you handle consent and privacy, and what would you deliberately exclude from a v1?

Practice this question out loud. An AI interviewer asks it, follows up like a real interviewer would, and scores your answer. Type or speak.

What this question tests

Product design balancing parental oversight, teen autonomy, and privacy in a sensitive, trust-dependent feature.

Define what parents get: visibility into usage patterns and topic categories at a high level, and the ability to set content and time restrictions, without full transcript access by default.
Define what teens retain: awareness that monitoring is active, never silent surveillance, and control over some settings within parent-set boundaries, preserving a degree of dignity and trust.
Handle consent explicitly: require verified parental consent to link accounts, with a clear, teen-visible explanation of exactly what is and is not shared.
Exclude from v1 deliberately: full conversation transcript sharing to parents, since that creates a strong disincentive for teens to use the product honestly for anything sensitive, undermining safety rather than helping it.
Define success: parent-reported comfort with the controls, and teen retention and honest usage, watched together, since a safety feature that drives teens away or toward workarounds is a failure.

Draws a clear, defensible line on transcript privacy, explaining that full visibility would likely drive teens to hide usage entirely, undermining the safety goal.
Proposes topic-level or pattern-level visibility as the middle ground, giving parents real signal without full surveillance.
Requires transparent, teen-visible disclosure of what is shared, treating trust as part of the safety design, not separate from it.
Names a dual success metric, parent comfort and teen honest engagement, showing awareness that over-indexing on either side undermines the other.

Proposing full transcript access for parents as a default, likely to push teens toward workarounds or alternative unsafe tools.
Ignoring teen-side transparency, designing monitoring that operates silently without the teen's knowledge.

What would you do if a parent demanded full transcript access despite the design tradeoff?
How would you handle a teen who creates a second, unmonitored account to bypass controls?