Legal and Compliance

Virginia ordered a statewide plan for AI-related cyber risks as the federal government pursues a lighter national approach and challenges some state rules.

Gerben IP reports that the USPTO denied Anthropic’s CLAUDE MYTHOS trademark application, but the decision’s date, grounds and status are not available.

Courthouse News reports a countersuit tied to an xAI forum must proceed against a woman described as Elon Musk’s former girlfriend; key details are unavailable.

OpenAI has a page titled “Our approach to EU text provenance rules,” but the available material does not reveal its claims, measures or timing.

A proposed CMMC Level 2 readiness workspace would turn self-assessment answers into draft compliance documents and a remediation plan.

The Hill reports that a DC appeals court sided with the Pentagon in a dispute over Anthropic’s blacklist designation; the ruling’s terms remain unclear.

A monitoring brief flags an EU Chat Control vote as a potential operations signal, but the available details do not establish its status or supply-chain effects.

IdeaNavigator AI outlines a narrow ’empty trust tracker’ SaaS for estate-planning firms and advisors to catch unfunded trusts before clients die.

OpenAI says Harvey is using GPT-6 Astra to draft from legal context, but the available announcement gives no performance or rollout details.

A proposed pilot would test phone-based parental consent records with 10 camps and youth programs; no deployment or results have been reported.