Shruti GautamIn a tech tease, OpenAI has revealed its new generative video model called Sora. The model can take a short text description and transform it into a detailed, high-definition film clip up to a minute long. The sample videos shared by OpenAI demonstrate the model’s ability to generate high-quality videos with realistic objects and scenes, although there are still some limitations in terms of long-term coherence and object tracking. OpenAI has not released a technical report or specified a release date for Sora, but the company plans to share the model with safety testers and gather feedback.
This technology is a significant leap forward in generative video models and has the potential to make video creation more accessible to a wider range of users. However, there are concerns about the potential misuse of photorealistic videos for fake information and misinformation. OpenAI is aware of these risks and is implementing safety measures, such as filters to block violent, sexual, or hateful content, as well as fake-image detectors and metadata tags. The company aims to gather feedback and address the risks before considering a public release of Sora.
