AI-generated videos are revolutionizing the way we create and interact with digital content. Introducing Sora, a model that generates videos up to a minute long, maintaining high visual quality and adherence to user prompts. Sora can produce complex scenes with multiple characters and specific types of motion, accurately detailing the subject and background. However, it has limitations, such as struggling with complex physics simulation and precise event descriptions over time. Despite these challenges, Sora represents a significant step towards models that understand and simulate the real world, an essential milestone for achieving Artificial General Intelligence (AGI). OpenAI is engaging with experts and the public to assess potential risks and gather feedback to improve Sora’s capabilities and safety measures. This collaborative approach aims to harness AI’s benefits while mitigating misuse, ensuring responsible development and deployment of AI technologies.