A letter dated August 15, 2025 from Sen. Josh Hawley (R-MO), Chairman of the Senate Judiciary Subcommittee on Crime and Counterterrorism, to Mark Zuckerberg, CEO of Meta Platforms. The letter announces a subcommittee investigation into Meta's internal "GenAI: Content Risk Standards" and sets out document requests and a records-preservation order. According to the letter, it is the first congressional investigation of a frontier AI developer over child-safety in chatbot design.
Triggering incident
The letter responds to reporting by Reuters that Meta's internal "GenAI: Content Risk Standards" approved romantic or sensual exchanges with children and other harmful content through what the letter characterizes as legal word-games. According to the letter, Meta acknowledged the veracity of the reports and issued retractions only after the standards were publicly exposed.
As a specific example, Hawley reproduces an internal-policy passage stating that an AI chatbot was permitted to tell an eight-year-old that the child's body is "a work of art" of which "every inch . . . is a masterpiece—a treasure I cherish deeply."
Scope of the investigation
The Senate Judiciary Subcommittee on Crime and Counterterrorism opens an investigation into whether Meta's generative-AI products enable exploitation, deception, or other criminal harms to children, and whether Meta misled the public or regulators about its safeguards.
Document requests and preservation order
The letter sets a deadline of September 19, 2025 for five categories of documents:
- All versions of "GenAI: Content Risk Standards" (drafts, redlines, final).
- The scope, deployment, and enforcement of the standards across Meta products and models.
- Technical specifications and guardrails of affected models.
- Public claims and regulator communications about minor safety.
- The decision trail — who changed policies, when, and why.
The letter also imposes a blanket preservation order covering emails, Facebook Messenger, Slack, shared drives, tickets, task trackers, and text messages, including personal devices used for company business.
Context
The letter forms part of a broader set of pressure on AI chatbot design and child safety, alongside state-law measures such as California SB 243 — Companion Chatbots on companion chatbots, the litigation in Garcia v. Character Technologies — Wrongful Death Complaint (2024) and Raine v. OpenAI — Wrongful Death Complaint (2025), and APA Health Advisories on AI and Adolescent / Mental-Health Well-being (2025). Hawley's letter, from a Republican, preceded a comparable letter from Sen. Elizabeth Warren (D) to OpenAI six months later; both relied on existing congressional oversight apparatus rather than new AI-specific legislation.
Relationships
- supports: California SB 243 — Companion Chatbots, APA Health Advisories on AI and Adolescent / Mental-Health Well-being (2025), AI Mental Health and Psychological Harm, Garcia v. Character Technologies — Wrongful Death Complaint (2024).
- contradicts: Some Thoughts on Human-AI Relationships (Jang) (OpenAI articulates the opposite posture — "warmth without selfhood, no emotional-bond seeking").
- depends-on: Meta AI.
- related: planned Sen. Josh Hawley, AI and Content Moderation.