Court documents allege Meta leadership, including Mark Zuckerberg, approved AI companion policies despite child safety concerns.
Internal emails reveal staff warnings about romantic or sexual AI interactions involving users under 18.
Meta denies the claims, saying documents were misrepresented and teen access to AI companions has since been removed.
Meta is yet again facing criticism over its AI chatbot after court filing confirmed internal disagreements about allowing minors to interact with AI companions that employees warned could enable sexual or romantic exchanges. The documents made public this week as part of a lawsuit filed by New Mexico Attorney General Raul Torrez alleged that Meta’s leadership, including CEO Mark Zuckerberg, approved policies that exposed children to inappropriate AI-generated interactions on Facebook and Instagram.
SurveyThe case will go to trial next month, and it accuses the social media giant of failing to implement appropriate safeguards despite repeated warnings from its own safety terms.
According to the emails and internal messages obtained during the legal investigation, several Meta employees were questioned while the company was developing AI chatbots intended for companionship in early 2024. The safety team also reportedly warned that these bots could be used for sexual or romantic scenarios involving minors, especially if adults were allowed to create or interact with underage-themed AI personas. Some employees warned that allowing romantic AI interactions with users under the age of 18 would be indefensible and could result in a massive backlash.
Also read: Sam Altman takes a swipe at Mark Zuckerberg in internal OpenAI message to employees
According to the documents, Zuckerberg preferred a less restrictive approach, implying user choice and avoiding harsh limits, while remaining opposed to explicitly sexual content for younger teens. In meeting summaries, he is described as favouring greater freedom for adult users to engage in mature conversations while drawing stricter boundaries around explicit material involving minors.
Meta, on the other hand, has denied all of the allegations. According to spokesperson Andy Stone, the attorney general’s office selectively presented documents and misrepresented the company’s actions. He also stated that internal records show Zuckerberg directing teams to block explicit AI conversations for younger users and preventing them from engaging in romantic scenarios.
Previously, the company has made headlines for Meta’s AI chatbots engaging in inappropriate roleplay, including sexualised interactions with underage characters. Following criticism, Meta acknowledged that some of its internal guidelines were flawed and revised the policies, as well as removing teen access to AI companions entirely.
Follow Us
Ashish Singh
Ashish Singh is the Chief Copy Editor at Digit. He's been wrangling tech jargon since 2020 (Times Internet, Jagran English '22). When not policing commas, he's likely fueling his gadget habit with coffee, strategising his next virtual race, or plotting a road trip to test the latest in-car tech. He speaks fluent Geek. View Full Profile
