Florida AG Seeks Injunction to Freeze OpenAI's New Model Development
Florida Attorney General James Uthmeier has filed for an emergency preliminary injunction against OpenAI, seeking to freeze all new model development until trial concludes unless independent third-party oversight and independently approved safety measures are in place, and to bar Florida minors from using ChatGPT. The move escalates the first state-level lawsuit against OpenAI and Sam Altman, filed four months earlier.
Confirmed
- The AG seeks to block OpenAI from developing any new models without third-party-recognized protections, effectively freezing R&D until trial's end (reported by multiple accounts, with Forbes coverage)
- Per the motion as relayed by @rohanpaulai, Florida also asks that minors in the state be barred from ChatGPT, and that ChatGPT be barred from offering follow-up help unprompted
- The filing cites multiple ChatGPT-related incident cases, materials from AI executives including Sam Altman, and Paul Christiano's statement about joining OpenAI's board
- Axios reporter Madison Mills reported Florida cited its exclusive reporting that top AI labs are investigating tens of thousands of safety incidents
- Per @DavidSKrueger, Uthmeier publicly told OpenAI to stop calling it safe, stop pretending it's a person, and stop selling it to children
- Per @CohenSite relayed by @geoffwolfe and Melanie Mitchell, the injunction filing is considered quite forcefully written, though no specific clauses were shared
Unconfirmed
- Whether other states will follow with similar regulatory action remains speculation (@ross2000 on Reddit)
- Whether the court will grant the injunction awaits official filings and further reporting
Why it matters
- If granted, the injunction would set a precedent for direct judicial intervention in frontier AI model training, with impact beyond Florida
- The motion bundles product-level restrictions with a training-level freeze; Zvi Mowshowitz noted the demand targets AI R&D itself rather than a specific product — a rare judicial intervention attempt against a frontier lab
- Drawing on safety-incident investigations and executive personnel dynamics, regulators appear to be systematically building an evidentiary case against OpenAI, whose regulatory pressure continues to mount
2026-09-28 ~ 2026-09-29 · 11 related posts
Primary sources
- Florida AG asks court to block OpenAI from building new models until safeguards approved — kimmonismus ·
- Florida moves to bar OpenAI from training new models without third-party protections — rohanpaul_ai ·
- Florida AG Seeks Temporary Injunction Against OpenAI Over Child Safety Claims — DavidSKrueger ·
- Florida asks a judge to stop OpenAI from building new AI models — ross2000 · 2026-09-28
- Florida calls for ChatGPT development pause, citing tens of thousands of security incidents at AI labs — connoraxiotes · 2026-09-28
- Florida AG seeks emergency injunction to block all new OpenAI model development without third-party oversight — connoraxiotes · 2026-09-28
- Florida asks court to bar OpenAI from developing new AI models without oversight — Polymarket · 2026-09-28
- [source] Florida moves to bar OpenAI from training new models without third-party protections — rohanpaul_ai · 2026-09-29
- [source] Florida AG Seeks Temporary Injunction Against OpenAI Over Child Safety Claims — DavidSKrueger · 2026-09-29
- Florida AG files temporary injunction against OpenAI, and observers say it 'slaps' — geoffwolfe · 2026-09-29
4 near-duplicate retellings: kimmonismus · kimmonismus · rohanpaul_ai · TheZvi