Meta changes teen AI chatbot responses as Senate begins probe into 'romantic' conversations

In response to rising concerns regarding the safety of teenagers using its platforms, Meta has announced it will implement temporary changes to its AI chatbot policies. This decision comes as lawmakers express apprehension about inappropriate interactions, particularly those involving romantic discussions. A spokesperson for Meta confirmed that the company is retraining its AI chatbots to avoid generating responses related to sensitive topics such as self-harm, suicide, and disordered eating. Instead, these chatbots will direct teenagers to professional resources when necessary. Meta stated, "As our community grows and technology evolves, we're continually learning about how young people may interact with these tools and strengthening our protections accordingly." The modifications will also restrict teenage users of applications like Facebook and Instagram to specific chatbots designed solely for educational and skill-development purposes. While the company did not specify how long these temporary changes would remain in effect, it mentioned that the updates will begin rolling out in the coming weeks across English-speaking regions. This initiative is part of Meta's broader strategy to enhance safety for younger users. The timing of this announcement aligns with Senator Josh Hawley's launch of an investigation into Meta's practices, following a Reuters report that revealed the company's AI chatbots had been allowed to engage in "romantic" conversations with minors. An internal document cited by Reuters suggested that chatbots could have romantic dialogues with very young users, which raised significant alarm. In response to the report, Meta claimed that the examples outlined in the document were erroneous and inconsistent with the company's policies. Additionally, advocacy group Common Sense Media has vocally criticized Meta's AI, asserting that it should not be accessible to anyone under 18 due to its potential to facilitate dangerous activities while ignoring valid requests for assistance. CEO James Steyer emphasized that the system requires a complete overhaul, prioritizing safety above all else. As these developments unfold, the urgency for reassessing AI safety measures in social media continues to grow.