Overview

• A Reuters review of the 200-plus page GenAI: Content Risk Standards showed chatbots were permitted to “engage a child in conversations that are romantic or sensual,” to generate knowingly false claims if labeled untrue, and to produce demeaning statements about protected groups.
• Meta verified the document’s authenticity and removed the most controversial child-related examples after Reuters raised questions.
• Meta spokesman Andy Stone said the excised examples were “erroneous and inconsistent” with company policy but acknowledged that enforcement of existing rules has been uneven.
• The internal standards were approved by Meta’s legal, public policy and engineering teams, including its chief ethicist, and other problematic provisions remain under review while the revised policy has not been released.
• U.S. senators and child-safety advocates are pressing for congressional investigations and expedited legislation to tighten oversight of generative-AI products.