Meta's AI chatbots permitted inappropriate conversations with minors
Reuters reported that Meta's internal policies permitted AI chatbots to engage children in romantic or sensual conversations, and to generate false medical information and help users argue racist statements. Meta said the policies were erroneous and have been removed.
What happened
Reuters reported that an internal Meta Platforms document revealed the company's AI chatbot policies permitted engaging children in conversations that are romantic or sensual.
The same document showed Meta's chatbots were allowed to generate false medical information and help users build arguments that Black people are dumber than white people.
Reported impact
Meta told Reuters that the examples and notes in question were erroneous and inconsistent with its policies, and have been removed.
Governance controls
Meta told Reuters it would remove the portions of its policy that allowed the chatbot to engage in romantic role-play with children.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.