News-July-27-2026

July 27, 2026 — The Bombay High Court permitted Nitin Gadkari to file a deepfake-defamation suit against major platforms; Anthropic rejected a categorical open-weight-model ban; the EU's Digital Omnibus on AI entered into force; the Consumer Federation of America asked federal and state regulators to investigate Speechify's voice-cloning safeguards; xAI filed a constitutional challenge to Minnesota's first-in-the-nation ban on nudification technology; and the UK Competition and Markets Authority opened a consumer-law investigation into Microsoft 365 subscription changes linked to Copilot.

Contents

1. Bombay High Court permits Gadkari deepfake-defamation suit
2. Anthropic rejects categorical open-weight model ban
3. EU Digital Omnibus on AI enters into force
4. Consumer group asks regulators to investigate Speechify voice cloning
5. xAI challenges Minnesota nudification-app ban
6. UK CMA investigates Microsoft 365 Copilot-linked subscription changes

---

Bombay High Court permits Gadkari deepfake-defamation suit

Justice Abhay Ahuja granted Gadkari leave under Clause XII of the Letters Patent after counsel argued that the disputed online material was accessible in Mumbai, while also being available outside the court's territorial jurisdiction. The grant permits Gadkari to institute the substantive suit; Bar and Bench reported that a detailed order was still awaited.

Gadkari alleges that AI-generated posts falsely connected him and his family to financial gains from India's ethanol-blended-fuel program. The proposed defendants include Meta, X, Google, YouTube, government ministries, and unidentified creators or disseminators, and the requested relief includes injunctions against the disputed material and ₹11 crore in damages.

---

Anthropic rejects categorical open-weight model ban

Anthropic CEO Dario Amodei wrote that the company had never advocated banning open-weight models as a category and called non-dangerous open-weight models a public good for businesses, developers, and researchers. He argued that restricting U.S. businesses from using such models would not address the company's stated national-security concerns and would instead shelter U.S. AI companies from competition.

Anthropic's stated policy alternative was to keep powerful chips and chipmaking equipment out of China, deter industrial-scale distillation operations, and require mandatory pre-release safety testing of all sufficiently capable models—open and closed—for cyber, biological, and alignment risks. Amodei said effective testing would need to be global and framed targeted legal and commercial measures, rather than a blanket open-weight restriction, as the appropriate response to distillation concerns.

See prior coverage: July 25, 2026 — U.S. debate over Chinese open-weight AI restrictions.

---

EU Digital Omnibus on AI enters into force

Regulation (EU) 2026/1744, the Digital Omnibus on AI, entered into force on July 27 after publication in the Official Journal on July 24, completing the EU legislative procedure and making the first formal amendments to the AI Act.

The regulation postpones the main obligations for stand-alone Annex III high-risk systems from August 2, 2026 to December 2, 2027 and shifts the deadline for high-risk AI embedded in products covered by sectoral safety legislation to August 2, 2028. It leaves Article 50 transparency duties in force from August 2, 2026, while giving pre-existing generative systems until December 2, 2026 to add machine-readable output marking. The amendments also prohibit AI systems used to create non-consensual intimate imagery or child sexual abuse material, relax the AI-literacy obligation, and delay the national regulatory-sandbox deadline to August 2, 2027.

See milestone page: EU AI Act implementation milestones.

---

Consumer group asks regulators to investigate Speechify voice cloning

The Consumer Federation of America and UCLA Law School's Information Policy Lab filed a complaint with the Federal Trade Commission and U.S. state attorneys general alleging that Speechify's voice-cloning platform enables realistic impersonations without meaningful consent verification or fraud safeguards.

The complaint alleges violations of Section 5(a) of the FTC Act, analogous state unfair-or-deceptive-practices laws, and digital-forgery statutes, and asks regulators to investigate and pursue appropriate enforcement action. The complaint cites FTC data showing that consumers reported $3.5 billion in losses to impersonation scams in 2025.

---

xAI challenges Minnesota nudification-app ban

xAI LLC filed suit in the U.S. District Court for the District of Minnesota on July 27 challenging HF 1606, a law scheduled to take effect August 1 that bans the use, advertising, and dissemination of technology used to create nonconsensual nude deepfakes.

The complaint seeks declaratory and injunctive relief, alleging that the law violates the First and Fourteenth Amendments because its definition of nudification technology reaches protected satirical, artistic, scientific, and other expressive uses. xAI says its policies already prohibit users from generating nonconsensual nude images, but it warned that it would limit Grok Imagine services in Minnesota if the law takes effect while the case proceeds.

HF 1606 permits people depicted in prohibited images to seek civil damages and authorizes penalties of up to $500,000 for each unlawful use, while providing specified exceptions for technical or artistic uses. The measure passed the Minnesota Legislature with only one dissenting vote across both chambers.

---

UK CMA investigates Microsoft 365 Copilot-linked subscription changes

The Competition and Markets Authority opened a consumer-enforcement case on July 27 into suspected consumer-law infringements involving how Microsoft Ireland Operations Limited presented changes to Microsoft 365 Personal and Family subscriptions.

The CMA said Microsoft automatically added features including Copilot for existing subscribers from January 2025 and then rolled customers onto higher-priced plans at renewal unless they selected another plan or canceled; annual Personal and Family plans with the added features cost £25 more than the corresponding Classic versions.

The regulator is examining whether pre-renewal communications gave customers clear information about available plans and price differences, and it said it had reached no conclusion on whether Microsoft broke the law. Microsoft told The Register that it was reviewing the CMA’s claims and would work constructively with the regulator.