OpenAI introduces its new video generation model, Sora

The OpenAI Sora text-to-video model, which can generate photorealistic videos up to one minute long based on user prompts, brings technology and creativity together.

WHAT ARE SORA'S FEATURES?

Sora allows for a wide range of creativity, including complex scenes with multiple characters, specific types of motion, and accurate details of the subject and background. According to OpenAI's introductory blog post, the model can understand how objects exist in the physical world, interpret props accurately, and generate compelling characters that express vivid emotions.

Demos created with Sora include an aerial shot of California during the gold rush, a video that looks like a train ride through Tokyo, and others. However, OpenAI notes that the model may struggle to accurately simulate the physics of a complex scene.

Recently, while companies like Runway and Pika have drawn attention with their own text-to-video models, Google's Lumiere also stands out as one of OpenAI's primary competitors in this field. Similar to Sora, Lumiere offers users text-to-video generation tools and also provides the ability to create video from a static image.

For now, OpenAI Sora has been made available to "red team" members who are evaluating the model's potential harms and risks. On the other hand, OpenAI is also providing access to some visual artists, designers, and filmmakers to gather feedback. The company notes that the current model may not accurately recreate the physical laws of a complex scene and may not correctly interpret certain cause-and-effect instances.

Like other AI products, OpenAI will have to deal with the consequences of AI-generated videos being mistaken for reality. In this context, the company stated earlier this month that it had added watermarks to its text-to-image tool DALL-E 3, but that these could be "easily removed." Sora brings with it significant ethical and societal debates, not just in the worlds of technology and art.

News Source: 12punto