OpenAI, the organization behind ChatGPT and DALL-E, has revealed that it is testing a text-to-video model called Sora. This model aims to enable users to create realistic videos with a simple prompt. According to OpenAI, Sora can generate videos up to a minute long while maintaining visual quality and adhering to the user’s input. The platform can even take an existing still image and generate a video from it.
OpenAI CEO Sam Altman mentioned that access to Sora is currently limited to a number of creators during the testing phase. Users are invited to suggest prompts, and Altman has shared convincing results on the platform, showcasing videos generated based on these prompts.
Also Read: OpenAI Researcher Andrej Karpathy Departs
However, OpenAI acknowledges that the current model has weaknesses, such as occasional confusion between left and right or difficulties in maintaining visual continuity throughout the length of a video. The company emphasizes the importance of safety and plans to subject Sora to adversarial testing, involving users attempting to make the platform malfunction or produce inappropriate content.
OpenAI has expressed its commitment to engaging with policymakers, educators, and artists globally to address concerns and identify positive use cases for this emerging technology. It’s worth noting that other tech giants like Meta, Google, and Runway AI are also working on text-to-video AI technology and have showcased similar samples of their work.
