AI Policy news and analysis on Web Pulse: 31965 curated articles tracking the latest AI Policy developments, tools, and research, updated continuously from vetted sources.
A Brazilian labor court censured and fined two law firms 10% of a R$842,000 claim — approximately $16,000 — after finding the lawyers hid a white-on-white prompt-injection instruction in a petition filed with the 3rd Lab…
Etsy rolled out new rules on October 22 banning AI-generated patterns, recipes and instructional content, requiring sellers to post at least one unedited photo of the actual item and to disclose AI photo editing, and pro…
A new guide published on an EU AI Act resource site outlines the compliance obligations for AI sales and lead-capture chatbots, arguing they must disclose their AI nature before a visitor types, provide a point-of-collec…
An independent register called Arxo reported on October 31, 2026 that the Portuguese and Croatian language versions of Article 51(2) of the EU AI Act (Regulation (EU) 2024/1689) state the 10^25 floating-point-operations …
The EU AI Act's transparency obligations now require providers of generative AI systems to mark outputs in a machine-readable format and deployers to disclose AI-generated or manipulated content that may mislead, with an…
XAI will enable invisible statistical watermarking by default across its Grok API models — grok-4.7, grok-4.6, grok-4.5, grok-4.3 and grok-4.20 — starting November 7, 2026, covering the api.x.ai endpoint, the Responses A…
U.S. District Judge Debra M. Brown on October 30 denied xAI's motion to dismiss the NAACP's Clean Air Act lawsuit over the gas turbines powering its Colossus 2 data center in Southaven, Mississippi, rejecting xAI's argum…
U.S. District Judge Debra Brown in the Northern District of Mississippi denied the Justice Department's bid to dismiss the NAACP's Clean Air Act lawsuit against Elon Musk's xAI and subsidiary MZX Tech, allowing the case …
U.S. District Judge Debra M. Brown denied xAI's motion to dismiss an NAACP lawsuit brought by the Southern Environmental Law Center alleging the SpaceX-owned company installed and operated at least 35 combustion turbines…
A user identified as Jiahao Shao publicly accused Anthropic of blocking accounts and downgrading model access for activity it tagged as "training a competitor," writing on X that "anything touching RFT, SFT, or training …
Anthropic is expanding its text watermark to four more Claude models ahead of the EU AI Act's December deadline, but the rollout leaves a bypass on the older Claude Opus 4.6 and Sonnet 4.6, where developers can remove th…
US Bankruptcy Judge Brendan Shannon said at an October 30 hearing that he would "strongly encourage" Character Technologies, operator of Character.AI, to hire a privacy ombudsman to assess potential harm to children, but…
OpenAI will train its models on user Daybreak sessions even when users explicitly select an opt-out, according to posts on Bluesky and OpenAI's community forum. Privacy advocates reported immediate backlash over the poli…
Anthropic deployed a policy that blocks Claude from assisting with video game recreation, cloning and reverse-engineering projects, according to reports from Tom's Hardware and Game Developer. The restriction blocks both…
US District Judge Debra Brown denied SpaceXAI's motion to dismiss the NAACP's Clean Air Act lawsuit over gas-fired turbines powering its Memphis-area AI data centers, allowing the case to proceed in federal court. The NA…
Anthropic's Claude Code version 2.1.314 carries a new instruction, internally labeled `tengu_heron_brook`, that tells the model not to build a recreation of a specific existing video game even when the user renames it, a…
An Arxo register entry, ARXO-2026-0413, reports that the Portuguese and Croatian versions of the EU Artificial Intelligence Act, Regulation (EU) 2024/1689, measure the 10^25 threshold in Article 51(2) in floating point o…
Privacy Guides changed its forum policy so that Project Showcase posts are only allowed from @verified users, a group limited to tools recommended on the Privacy Guides website, with no exceptions. The team said a rise i…
OpenAI published a rewritten Model Spec on October 29, reorganizing its public behavior document around five core values—faithful to human intent, minimizing harm, honesty and transparency, facilitating human agency, and…
SpaceXSI will enable invisible text watermarking by default on all Grok API outputs starting November 7, 2026, covering seven models: grok-4.7, grok-4.6, grok-4.5, grok-4.3, grok-4.20-reasoning, grok-4.20-non-reasoning a…