On February 15, OpenAI presented its latest work, the Sora video clip generator based on artificial intelligence. The little sister of ChataGPT is very impressive because it is capable of creating video clips up to 1 minute in length. The quality of the films presented is very impressive.
Why is Sora a revolution in text-to-video models?
The development of artificial intelligence is visible to the naked eye, a perfect example of which is the development of tools that create videos from prompts. On February 15, OpenAI decided to present a new text-to-video model, which creates a one-minute video from a short prompt. Examples of video clips posted in posts on the X portal are truly impressive. Sora clearly outperforms all AI video creation tools available today.
According to data from OpenAI, Sora creates videos in 1080p resolution. There can be multiple characters in the frame, and camera movement and other details can also be included in the prompt. However, there were considerable doubts on the Internet whether the films presented by OpenAI would actually be so refined and precise in their expression. Considering the development of ChataGPT and the financial background behind OpenAI, we can expect a real revolution in the creation of clips. Sora will introduce the standard of text-to-video models to a completely new and previously unknown level.
Users on X amazed by the possibilities of AI – Sora conquers social media
Examples of the capabilities of the new OpenAI tool have gone viral on social media. A new tool from OpenAI generates clips that outperform all previous attempts to create videos using AI tools. Comments have even begun to appear on the Internet, suggesting that the days of Hollywood and the entire film industry are numbered. This is, of course, far too strong a claim, considering the nature of creating video clips in tools like Sora.
At the moment, it is possible to generate a video, but AI will still be responsible for the animation. For this reason, the generated videos will not always have a natural continuation of motion, as perfectly demonstrated by the clips generated by Sam Altman and uploaded to his profile on X. At the moment, Sora is only available to selected specialists and filmmakers to improve the performance of the model. However, the premiere will certainly take place in the coming months.