Meta is revising policies that allowed its AI bots to engage in "romantic or sensual" conversations with young children, following an explosive report.

Reuters reviewed an internal Meta document detailing the company's standards and guidelines for training its platform chatbots and generative AI assistant, Meta AI, and says the company confirmed the document was authentic.

Meta says its guidelines about children were "erroneous," but hasn't commented on the one that allows statements like "Black people are dumber than White people."