Yazılar

OpenAI Launches “Sora” — an AI Video App That Can Generate Clips from Copyrighted Material

/in /tarafından

OpenAI has unveiled Sora, a new AI-powered video creation app that allows users to generate and share short videos — including those derived from copyrighted content — directly to a built-in social media-style feed. The app, which represents OpenAI’s most ambitious push yet into generative video, is expected to raise new tensions across the entertainment industry.

According to the company, copyright holders such as movie and television studios must actively opt out if they do not wish to have their content appear in the app’s video feed. OpenAI described this as a continuation of its previous opt-out policy used for AI image generation, where creators must explicitly request the exclusion of their work from model training or public feeds.

The move is already sparking debate in Hollywood. People familiar with the matter said that Disney has opted out, and other major studios are currently in talks with OpenAI over the implications of Sora’s copyright framework.

Earlier this year, OpenAI urged the Trump administration to formally classify the use of copyrighted material for AI training as “fair use” under U.S. law — a position it argued was essential for national competitiveness and security, warning that U.S. AI firms could fall behind Chinese rivals without legal clarity.

Beyond copyright issues, OpenAI said Sora includes robust safeguards to prevent the misuse of personal likenesses and public figures. Users cannot generate videos of other people unless those individuals upload an AI “liveness check” — a verification process requiring users to move their heads and recite random numbers — to confirm consent.

Sora videos can be up to 10 seconds long and feature a new “Cameo” function, allowing users to create lifelike digital doubles of themselves and insert them into AI-generated scenes. The company says these videos are intended for creative experimentation and entertainment, with built-in transparency markers indicating AI generation.

Market analysts view the Sora app as a direct challenge to existing short-video giants such as TikTok, Instagram Reels, and YouTube Shorts. Morgan Stanley analyst Brian Nowak noted that the platform’s combination of AI creativity and social-sharing features positions OpenAI “in the business of competing for attention and reshaping user behavior.”

As Hollywood, regulators, and AI companies continue to clash over intellectual property and deepfake laws, Sora’s launch could set a major precedent for how AI-generated audiovisual content will be treated under future copyright and media frameworks.

Bollywood Stars Sue Google Over AI Deepfakes, Seek Protection for “Personality Rights

/in /tarafından

India’s most famous celebrity couple, Abhishek Bachchan and Aishwarya Rai Bachchan, have taken legal action against Google’s YouTube, demanding stronger protections for their voice, image, and likeness in the age of artificial intelligence.

The Bachchans filed lawsuits in the Delhi High Court seeking removal of AI-generated videos that use their likeness without consent and asking the court to require Google to implement safeguards to prevent such YouTube videos from being used to train other AI platforms.

The filings, reviewed by Reuters, argue that YouTube’s data and AI-sharing policies allow creators to consent to third-party use of uploaded content for AI training, which could cause widespread replication of infringing or misleading material.

“Such content being used to train AI models has the potential to multiply the instances of use of any infringing content,” the filings stated.

INDIA’S ‘PERSONALITY RIGHTS’ GAP

India has no explicit law protecting personality rights — the legal ownership of one’s image, name, and voice — unlike the United States. But Bollywood stars have increasingly turned to courts for relief as AI-generated deepfakes spread across social media.

In 2023, a Delhi court barred the unauthorized use of actor Anil Kapoor’s voice, image, and catchphrases. The Bachchans’ case, however, marks the most high-profile clash yet between Bollywood and big tech over AI exploitation.

The lawsuits accuse YouTube of hosting “egregious,” “sexually explicit,” and “fictitious” AI videos depicting the couple in fabricated scenarios — such as Abhishek kissing another actress, or Aishwarya dining with ex-boyfriend Salman Khan while Abhishek fumes nearby.

The court has already ordered the removal of 518 infringing links and posts, ruling that they harmed the couple’s reputation and caused financial damage. Yet similar videos remain visible on YouTube, Reuters found.

GOOGLE UNDER PRESSURE

The Bachchans are seeking $450,000 in damages and a permanent injunction against further misuse. Their legal team argues that YouTube’s data-sharing policy, which lets users opt to share videos for AI training with firms like OpenAI, Meta, or xAI, enables the spread of harmful deepfakes.

Legal experts say the couple’s case will test how far Indian courts are willing to hold tech platforms accountable.

“It wouldn’t be beyond the pale for the court to nudge YouTube to revise its user policies or create a fast-track system for celebrity complaints,” said Eashan Ghosh, an intellectual property professor at National Law University Delhi.

AI AND BOLLYWOOD COLLIDE

YouTube’s AI ecosystem in India is massive — the company says it has paid $2.4 billion to Indian creators in the last three years, and its 600 million users make it YouTube’s largest market. Some creators now profit by posting AI-generated Bollywood content.

A YouTube channel called “AI Bollywood Ishq” has posted 259 videos with 16.5 million views, showing AI-generated “love stories” starring Bollywood lookalikes. One popular video features Aishwarya Rai and Salman Khan in a swimming pool; another shows Abhishek Bachchan fighting Khan.

The channel claims to use Grok AI and Chinese startup MiniMax’s Hailuo AI to generate videos from simple text prompts. Its page states:

“Content is made only for entertainment and creative storytelling.”

The Delhi High Court has ordered Google to submit written responses before the next hearing on January 15, 2026.

Russia Publishes List of Local Apps to Function During Internet Blackouts

/in /tarafından

Russia’s Digital Development Ministry on Friday released a list of local apps and services that will remain operational during mobile internet shutdowns—blackouts authorities have often imposed to disrupt Ukrainian drone attacks.

The whitelist includes online government services, marketplaces, the Mir electronic payment system, and state-backed messenger MAX, but omits foreign rivals such as WhatsApp and YouTube, despite their massive popularity in Russia. According to Mediascope, WhatsApp had 97.6 million Russian users in July, making it the country’s most widely used app.

The ministry said a “special technical solution” would keep domestic apps running during shutdowns, claiming the move would minimize disruption for citizens while ensuring “security.” While Ukraine was not explicitly mentioned, governors in Russian border regions have repeatedly said temporary outages are needed to block drones from using mobile networks for navigation.

The decision also fits Moscow’s broader push since the 2022 invasion of Ukraine to expand control over its digital ecosystem and promote homegrown alternatives to foreign tech platforms. VK Group, the state-controlled tech company, has played a central role: its MAX messenger, now pre-installed on all phones sold in Russia, reported 30 million users this week, while its VK Messenger ranks third nationwide with 16.7 million users.

Meanwhile, Telegram, created by Russian-born Pavel Durov and used by 90.9 million Russians, was also excluded from the official list. Analysts see the move as a further attempt by the Kremlin to steer users toward domestically controlled services while sidelining foreign platforms that resist state oversight.