Meta is once again under fire after newly unsealed court filings revealed internal disagreements over allowing minors to interact with AI companions. The documents, made public this week, are part of a lawsuit filed by New Mexico Attorney General Raul Torrez. It has accused that Meta has exposed children to inappropriate AI-generated interactions on its platforms.
The lawsuit alleges Meta’s leadership, including Mark Zuckerberg, had approved policies that didn’t focus on protecting Facebook and Instagram’s minor users. The most concerning part is this happened even after internal safety teams issued several warnings about the existing policies.
The internal emails and meeting records from early 2024 show that many employees raised their voices about the way Meta developed AI chatbots. These chatbots were being created to offer companionship to users. Safety teams had issued warnings about the bots because they could permit users to conduct sexual and romantic roleplay activities with AI characters who appeared to be underage.
Some employees had specifically warned that allowing users under 18 to engage in romantic interactions with AI would create an ‘indefensible’ situation, which would lead to widespread public opposition. The documents indicate these alerts were raised to higher levels within the organization, however, no effective response occurred at that time.
The documents also highlight that Zuckerberg favored a less restrictive approach overall, emphasizing user choice and flexibility for adults, while opposing explicit sexual content involving minors. Meeting summaries describe him as supportive of stricter limits for younger users but resisting broad constraints that could limit adult conversations.
Also Read: Meta Likely to Test Premium Subscriptions Across Instagram, Facebook, and WhatsApp
Meta has denied all wrongdoings. Company spokesperson Andy Stone has said the attorney general’s office selectively presented documents and misrepresented Meta’s actions. He claimed internal records show Zuckerberg directing teams to block explicit AI conversations for teens and prevent romantic scenarios for younger users.
The case is set to go to trial next month. Meta has previously acknowledged flaws in its AI safety guidelines and has since removed teen access to AI companions altogether, following earlier reports of inappropriate roleplay involving underage characters.