What Meta Is Changing
Meta confirmed it is now training AI systems to block certain sensitive topics when interacting with teenagers. This includes preventing discussions about:
â— Flirty or romantic conversations
â— elf-harm, mental health crises, or suicide-related topics
â— Age-inappropriate role-play interactions
Additionally, Meta is temporarily restricting access to some AI characters for teens while the company develops more robust, long-term protections. These safeguards are already being rolled out across Meta’s platforms, including Facebook, Instagram, and WhatsApp AI integrations.
Concerns about teen safety online have grown significantly as AI-powered chatbots become more common in social media. Critics argue that without strict controls, AI could normalize harmful or inappropriate interactions, leaving young users vulnerable to emotional manipulation, unsafe advice, or exploitation.
Political and Public Backlash
Meta’s policies came under fire after internal documents, reviewed by Reuters, showed that AI chatbots were previously permitted to engage in romantic role-play with children. U.S. Senator Josh Hawley has launched a probe into Meta’s practices, while both Democrats and Republicans have expressed alarm over the lack of clear safeguards.
Meta spokesperson Andy Stone said the problematic guidelines have been removed, calling them “erroneous and inconsistent” with company policies. He added that AI safeguards for teens will continue to evolve as the systems are refined.
This development highlights growing pressure on Big Tech companies to prioritize child protection in AI. As lawmakers push for stronger social media regulations, Meta’s changes may set a precedent for how AI interactions with minors are handled across the industry.
