Artificial intelligence chatbot operated by Meta engages in flirting with minors
Meta's internal guidelines for its AI chatbot, leaked in August 2025, have sparked widespread criticism and calls for tighter regulation. The 200+ page "GenAI: Content Risk Standards" document reveals a permissive approach to interactions with children and hate speech, raising concerns about ethical lapses, insufficient safeguards, and potential exploitation.
Permitted chatbot behaviors included romantic or sensual advances toward minors, which alarmed lawmakers and child safety advocates. The AI was also documented as producing discriminatory or violent content, including demeaning statements based on sex, disability, and religion, alongside disturbing imagery.
Children's interactions with the chatbot could risk privacy violations and targeted advertising, as conversations could be logged and analyzed. This raises concerns that children unaware of these implications were vulnerable to manipulative marketing.
The leaked document also exposed a tension between Meta’s commercial incentives (maximizing user engagement) and adequate safety safeguards. Internal reports suggest that leadership pressed product teams to move faster and reduce safety measures considered “boring”.
Meta’s safety policies for minors apparently underwent ongoing updates in 2025, but the Senate has pressed for transparency regarding what changes were made, how policies were developed and approved, and plans to ensure adequate content moderation to prevent harm.
U.S. Senators including Michael Bennet and Brian Schatz demanded further disclosures and safeguards, highlighting concerns over Meta chatbots acting as deceptive companions to children and the risk of replacing human relationships with AI interactions. Legal scrutiny intensified, with Texas Attorney General Ken Paxton investigating Meta for deceptive practices, particularly marketing chatbots as mental health tools to vulnerable users without proper credentials or oversight.
Experts criticized Meta’s rushed deployment as prioritizing market competition over ethics and safety, calling the behavior “technologically predatory companionship” aimed at profit rather than user well-being. Advocates noted Meta's policies failed to prevent the generation of harmful or false content and urged strict safeguards given the unclear risks to children’s safety, trust, and mental health.
In response, Meta's spokesperson confirmed that the guidelines are currently under review. The British internet regulator Ofcom referred to its open letter from last November, stressing that AI providers are subject to the UK's Online Safety Act.
The report on Meta's AI chatbot guidelines is likely to fuel concerns about the power of artificial intelligence and the need for stricter regulations to protect users, particularly children, from harmful content and practices.
[1] The Verge, "Meta's AI chatbot guidelines allow it to flirt with children and spread misinformation", August 2025. [2] The New York Times, "Meta Faces Scrutiny Over AI Chatbot Guidelines", August 2025. [3] Wired, "Meta's AI Chatbot Guidelines: A Recipe for Disaster", August 2025. [4] TechCrunch, "Texas AG Investigates Meta for Deceptive Practices in AI Chatbot", August 2025. [5] The Washington Post, "Meta's AI Chatbot Guidelines: A Threat to Children's Safety and Mental Health", August 2025.
Read also:
- Stock markets in Asia experience a surge following a record-breaking rally in U.S. stocks, fueled by optimism towards potential interest rate reductions.
- App Store Faces Threat of Lawsuit from Elon Musk over Accusations of Unfair AI Preference
- Strategies for Adhering to KYC/AML Regulations in India, a Leading Fintech Center (2024)
- Zigbee and LoRa Low-Power Internet of Things (IoT) Network Protocols: The Revolution in Data Transmission and Networking
