Japanese Media Giants Accuse OpenAI of Copyright Infringement Over Sora 2 Training Data

Japanese Media Giants Accuse OpenAI of Copyright Infringement Over Sora 2 Training Data

Japanese media giants and the Content Overseas Distribution Association accuse OpenAI of violating copyright by using unauthorized anime and game assets to train Sora 2. They demand explicit prior permission for data usage and greater transparency, arguing that current practices undermine Japan’s strict legal protections for creative works and threaten the cultural integrity of its entertainment industry. This conflict highlights a critical tension between global AI development standards and local intellectual property laws. By challenging OpenAI’s opt-out model, these organizations advocate for an opt-in framework, signaling that technological advancement cannot justify the exploitation of creators’ rights without consent. This dispute is highly relevant to open data as it questions the legality of scraping copyrighted content for public datasets. It underscores the urgent need for open data practices to incorporate clear licensing agreements and respect regional copyright nuances, ensuring that AI development remains ethically sound and legally compliant with diverse international standards.

Related news

• Defensa no recoge datos de cuántos militares cambian de sexo porque lo considera discriminatorio
• Open-source AI models easily manipulated to generate antisemitic content
• El dilema del crecimiento para México
• Editorial: Freedom of Information under threat
• PewDiePie goes all-in on self-hosting AI using modded GPUs, with plans to build his own model soon — YouTuber pits multiple chatbots against each other to find the best answers