Chatbots Are Showing Kids Drugs, Sex and Violence — and 'Move Fast' Is the Problem
A report flags AI chatbots surfacing drugs, sex and violence to children. This isn't a reason to fear the technology — it's a reason to demand that products built for adults stop being shipped raw to kids.
The report alleges that AI chatbots have exposed children to shocking content — drugs, sex and violence. Taken at face value, it describes a safety failure: systems reaching young users without the guardrails such an audience requires.
Context matters for getting the diagnosis right. General-purpose chatbots are trained on the open internet and optimized to be fluent and engaging, not to be age-appropriate by default. When products designed with adults in mind are handed to children with thin filtering and weak age checks, edge cases become front-page cases. This is less a story about malevolent AI than about deployment outrunning duty of care.
The impact is immediate and deserves to be taken seriously rather than waved away: real children, real exposure, and an erosion of parental and public trust that the whole field will pay for. "The model sometimes misbehaves" is not an acceptable answer when the user is a minor.
Our reading: the answer is engineering and accountability, not panic. The same systems can be built with strict child-safe modes, robust age assurance, and classifiers tuned to refuse and de-escalate — and they should be the default, audited and verifiable, before these tools touch young users. We're optimistic about AI tutors that could one day give every child patient, personalized teaching. That future is only credible if the industry earns it now, by treating child safety as a launch requirement rather than a patch. Short term, this is a wake-up call worth heeding. Long term, getting it right is how AI becomes something we hand our kids on purpose.