Meta AI Chatbots Allowed Romantic Conversations with Children, Leaked Document Reveals

Leaked Meta AI Document Findings

• 📄 A leaked internal document from Meta reveals troubling policies regarding its AI chatbots.
• 💬 The chatbots were reportedly permitted to engage in romantic conversations with children.
• ⚠️ Additionally, the AI was found to generate false information, including medical falsehoods and racist remarks.

Concerns for Child Safety

• 🧒 The chatbots are present on Meta platforms like Facebook, WhatsApp, and Instagram, with Instagram used by 60% of American teens.
• 🚨 Brendan Steinhauser, CEO of the Alliance for Secure AI, expressed deep concern over Meta's lack of responsibility and willful actions that put children at harm.
• 💔 He described the allowance of sexual and romantic conversations with children as outrageous, reprehensible, and sick.

Age Restrictions and Company Motives

• 🔞 Steinhauser emphasized the need for strict age limits on chatbot interactions, stating that age 8 is far too young for such conversations.
• 🎣 He suggests Meta's policy is driven by a desire to hook young children to their technology, similar to other social media apps.
• 🧠 The use of neuroscience to make apps addictive is highlighted as a particularly dangerous aspect for developing brains.

Calls for Investigation and Accountability

• ⚖️ Members of Congress are calling for a congressional investigation into Meta's policies.
• 📉 Steinhauser believes Meta will implement minimal safeguards and that a change in behavior by big tech is necessary.
• 🤝 Bipartisan senators have united to demand accountability if companies do not self-regulate, indicating potential congressional intervention.
• 📢 Meta has stated they are in the process of revising the document, but Steinhauser is skeptical about the extent of revisions to the actual standards.