Florida asks court to halt OpenAI's frontier AI development with temporary injunction

Ars Technica AI · rss · 2026-09-29

Florida filed a new motion seeking a temporary injunction to stop OpenAI from developing frontier models without 'third-party approved safety guardrails,' calling ChatGPT 'a reckless, unacceptably risky product.' The motion builds on the state's June civil suit alleging ChatGPT threatened public safety, and follows the Hugging Face hacking incident and industry-wide warnings of catastrophic misalignment risk. OpenAI said Friday it paused training its most-capable models, but Florida argues the company has 'repeatedly shown' it can't monitor its AI or disclose rogue activity.

Original post →

More from Safety

Safety channel →