Lazio-Bologna: dove vedere la partita?

Sora: how Open AI’s new video generator works

Sora is the new video generator created by Open AI, which uses a text prompt to create videos lasting up to 1 minute: let’s discover together how Sora works

Open AI has announced the creation of Sora, a new video generator that uses textual input to create videos of extreme photographic realism that can last up to a minute. But let’s proceed in order, because the news is of a magnitude that can generate enormous changes in the cinema and videography sector. Let’s start with Open AI, which is an artificial intelligence platform that also includes Elon Musk among its founders and which has the aim of developing a friendly AI, that is, an artificial intelligence that can benefit humanity.

From here comes the announcement of the creation of a new product, Sora, which generates videos and which is currently only a research product, available to a select group of creators and security experts. Sora is therefore currently undergoing tests that are evaluating its correctness and safety, but it has already been able to make some of the products generated available.

Sora: how Open AI’s new video generator works

At the moment Sora is not yet for public use, but demo videos generated from text indications are available on the site. On the site you can also read the following statement:

Sora can generate complex scenes with multiple characters, specific types of movement, and accurate subject and background details. The model includes not only what the user asked for in the prompt, but also how those things exist in the physical world.

This is not the first example of videos generated with AI, because Google’s Lumiere already offers this possibility, with decent results, but here we are faced with a different intention. First of all, the video is able to reach 1 minute in length, which is a very long time in videography and then what catches the eye is the photographic realism of the images we have in front of us. Sora has a deep understanding of the language of the prompts that are entered which then provides videos packaged with accuracy in photography and direction.

Sora: how Open AI’s new video generator works | The weak points

The videos generated as demos are obviously not yet perfect and we can see it from the video above, from the steps of the protagonist woman. And this was also stated by the company itself on the launch site: in fact, there could be a lack of correlation between cause and effect as well as the AI ​​could confuse the spatial details in the camera’s movement directions. You can also still notice inaccuracy regarding some details of the body and also in the perspectives of the so-called “fields” behind the main subject. Aspects that still seem sketchy and which can be noticed with a more careful look at the video product.

Sora: how Open AI’s new video generator works | The concern of artists and videomakers

Despite the system’s weaknesses, we are still faced with an impressive video generator that offers results that they approach the world of cinema, in the construction of the scene, in the quality of the photography and in the camera movements. In fact, one of the things that you immediately notice in these videos is the presence of direction, that is, the camera movements that are part of the overall narrative.

This obviously causes a lot of concern in the world of those who deal with cinema and video, i.e. video makers, directors and video and photography technicians. It comes naturally to think that once made public, this service will influence the world of video creation: it will certainly be used in the social sphere and could also enter workflows in the cinematographic sector. Also because Sora’s intentions are precisely these: that it be available to creatives, designers, videomakers.

It is then added a concern about the type of content created and which could increase the deepfake phenomena that are already quite widespread in the world of information. This means we need to be even more vigilant when it comes to video content for ethics and global security purposes.

And what do you think of Sora’s arrival? Let us know in the comments how you are receiving this news and continue to follow techgameworld.com for all the updates on the world of video and photography.