Recent Stories

The Justice Department announced it will appeal Judge Rita Lin's August 27 ruling that struck down the Pentagon's supply-chain risk designation of Anthropic as unconstitutional First Amendment retaliation, keeping the legal battle alive and raising the possibility the federal agency Claude ban could be reinstated pending appeal.

Within days of the EU AI Act's August 2 enforcement start, the AI Office confirmed it opened formal scrutiny of Anthropic and OpenAI under new general-purpose AI enforcement powers, marking the first use of the regulation's GPAI regime against frontier model providers.

AB 1883 (algorithmic management restrictions) and SB 947 (automated workplace decision protections) are at third reading and must pass California's full legislature by August 31 -- if signed by Governor Newsom, they would be the first US state laws restricting employer use of AI-driven management systems.

Judge Rita F. Lin found the DOD's supply-chain risk designation of Anthropic was illegal First Amendment retaliation and a Fifth Amendment due-process violation, ordering it removed and restoring Claude access for federal workers at agencies including HHS. A second Anthropic lawsuit remains pending in DC.

SB 503 (medical AI bias protections), AB 1883 (algorithmic management), and SB 947 (workplace AI decisions) are among 24 California AI bills heading to final floor votes; if signed by Governor Newsom, several would be first-in-the-nation worker and health AI protections.

AB 1883 (algorithmic management), SB 947 (AI workplace decisions), and SB 951 (AI-driven workforce reduction notice) headline 24 California AI bills that must clear both chambers by August 31 before heading to Governor Newsom -- if signed, the first US state laws restricting employer use of automated management systems.

Army CIO Leonel Garciga signed a memo on August 26 requiring all Army AI products to comply with records management, privacy protections, and FOIA -- including explicit disclosure when AI-generated content is responsive to public records requests.

SI 2026/425 (in force May 12, 2026) requires the UK Information Commissioner to produce a statutory AI and ADM code of practice; once finalized, failure to follow the code will constitute evidence of non-compliance in ICO enforcement actions and litigation, with the code expected in 2027.