News-August-12-2026

August 12, 2026 — German digital-rights group HateAid filed a criminal complaint over Meta AI smart glasses and German privacy law. The Delhi High Court granted actor Janhvi Kapoor interim protection against AI-generated impersonation, deepfake pornography, and other misuse of her personality, while declining to impose a blanket ban on legitimate fan pages and commentary. India’s Supreme Court directed the Union government to consider a request for a rapid, URL-specific process addressing non-consensual AI-generated content, deepfakes, doxing, and threats. Twitch also confirmed that creators’ channel content is available by default for training Amazon generative-AI models, while adding an account setting that lets creators opt out.

Contents

1. HateAid files German criminal complaint over Meta AI glasses
2. Delhi High Court protects Janhvi Kapoor from AI deepfakes
3. India’s Supreme Court asks government to consider rapid AI-content remedy
4. Twitch defaults creators into Amazon AI training

HateAid files German criminal complaint over Meta AI glasses

HateAid filed a criminal complaint with Frankfurt’s Central Office for Combating Cybercrime against Meta management, EssilorLuxottica units including Ray-Ban, and retailers Fielmann, Apollo-Optik, Mister Spex, and MediaMarkt.

The group alleged that the Ray-Ban Meta Wayfarer and related devices violate German privacy law because they can record people without sufficient notice. HateAid said its complaint relies on a federal data-protection provision prohibiting the sale of communication devices designed to record people without their knowledge.

Mister Spex said it had not been formally notified of the complaint and that it takes privacy protection seriously; the other named companies and the prosecution office did not immediately respond to Reuters requests for comment.

---

Delhi High Court protects Janhvi Kapoor from AI deepfakes

Justice Anup Jairam Bhambhani granted Janhvi Kapoor interim protection against specified online content allegedly using artificial intelligence to impersonate her, create deepfake pornography, or commercially exploit her name, image, and likeness without consent.

The court identified three categories suitable for protection: clearly pornographic or obscene content, content directly monetizing her personality, and content using her identity to sell goods or services. It declined to impose a blanket injunction on all online material referring to Kapoor, reasoning that fan pages may contain lawful satire, praise, criticism, or commentary.

The court directed Google to de-index identified URLs and ordered social-media intermediaries to remove or disable access to specified content within 72 hours after receiving the order.

---

India’s Supreme Court asks government to consider rapid AI-content remedy

A three-judge Supreme Court bench directed India’s Union government to consider advocate Narendra Kumar Goswami’s representation seeking an urgent reporting and URL-specific access-disabling mechanism for online threats, doxing, unauthorized disclosures, and non-consensual synthetic or AI-generated content.

The court asked the ministries of Electronics and Information Technology, Home Affairs, and Law and Justice to examine the representation and take remedial measures. The petition proposed preservation and judicial-review safeguards and said any mechanism should not authorize prior restraint or emergency takedowns of political speech, journalism, satire, or fair criticism.

---

Twitch defaults creators into Amazon AI training

Twitch said creators’ channel content may be used to train generative-AI content models across Amazon and introduced an account setting allowing creators to opt out; creators are included by default unless they change that setting.

Twitch Chief Product Officer Mike Minton said during an official Twitch livestream that the setting was not opt-in because “nobody would opt in.” He also said he did not know whether Amazon had already used a questioner’s videos for model training.