OpenAI releases stunning new Sora videos

OpenAI Sora clips suggest a promising future for true generative entertainment
An undated image of OpenAIs Sora text written on the cloud. — OpenAI
An undated image of OpenAI's Sora text written on the cloud. — OpenAI 

OpenAI continues to tease the capabilities of its Sora — a generative video model. The new clips show innovations that are much like the production of Hollywood and are all generated from a single prompt.

Sora isn’t accessible to anyone, it is available only to a limited group of testers within OpenAI, but netizens share glimpses of its potential social media platforms, offering insights into its capabilities.

The first couple of video releases featured scenes such as dogs playing in the snow, a couple in Tokyo, and a flyover of a 19th-century California gold mining town.

Now you can see the clips from a single AI prompt that look like a complete media production with multiple shots, effects and motion in up to a minute of video.

Here are some of the new clips

These glimpses suggest a promising future for true generative entertainment, particularly when combined with other AI models for sound, lip-syncing, or production platforms like LTX Studio, making creativity more accessible.

Blaine Brown, a user on X (formerly Twitter) shared a video that combined Sora Alien by Bill Peeble (A user on X former Twitter) with Pika Labs Lip Sync and music generated using Suno AI to create music videos.

How does Sora compare?

— Tomsguide
— Tomsguide

In comparison to other AI video models like Runway's Gen-2, Pika Labs Pika 1.0, and StabilityAI's Stable Video Diffusion 1.1, Sora can do things that are not possible with other models. Sora utilises a combination of the transformer technology in chatbots like ChatGPT with image generation diffusion models found in Midjourney, Stable Diffusion and DALL-E.

While existing AI video tools typically produce clips of 1 to 4 seconds and sometimes struggle with complex motion.

— Tomsguide
— Tomsguide

However, other AI companies are taking note of Sora's capabilities. StabilityAI plans to introduce Stable Diffusion 3 with a similar architecture, and Runway has enhanced its Gen-2 model for more consistent motion and character development. Additionally, Pika introduced Lip Sync to enhance character realism, according to Tomsguide.