OpenAI launched its new generative AI model, Sora, on Thursday, February 16, 2024. Sora is a text-to-video generator capable of creating photorealistic videos up to 60 seconds long, leveraging a diffusion model to understand and simulate the physical world in motion. As of now, Sora is not available for general public use, as OpenAI is working to improve its safety features, including rejecting text prompts that involve extreme violence, sexual content, hateful imagery, or infringing on third-party IP or celebrity privacy rights.
Sora aims to support people in solving real-world interaction problems and is currently being tested by red teamers, domain experts who will adversarially evaluate the model. The launch of Sora is a pivotal moment for text-to-video generation, showcasing advancements in AI capabilities and the potential for creative professionals.