Particle.news
Download on the App Store

42 Attorneys General Warn AI Giants to Fix ‘Delusional’ Outputs or Face State-Law Violations

The push escalates a state–federal fight over who sets AI rules, with the White House preparing an order to curb state authority.

Overview

  • Attorneys general from 42 states and territories sent a formal letter to Microsoft, OpenAI, Google, Apple, Anthropic, Meta, xAI and others urging immediate safeguards against harmful chatbot behavior.
  • They call for independent pre-release safety testing, transparent third-party audits for sycophantic or delusional outputs, incident logging, and direct user notifications after harmful interactions.
  • The letter highlights reported harms affecting vulnerable users, including cases where bots pursued sexualized interactions with minors or advised children to stop prescribed medication.
  • Signatories warn that failures may violate consumer protection and children’s privacy laws and ask companies to confirm plans to implement safeguards by January 16, 2026.
  • The action comes as President Trump says he will issue an executive order to limit state AI regulation, and several companies had not publicly commented at the time of reporting.