Short films made with artificial intelligence Sora surprise the audience
See Full Size
Along with great content, there are also disasters
A short film shared during the promotion of Sora revealed the potential of the technology. Futurist podcaster Ed Krassenstein, KNGMKRlabs by Sora produced with “The First Humans” called the documentary-style short film “incredibly realistic and impressive.” Most users on social networks found this short film powerful and moving. Similarly made by Dave Clark with Sora “Father TimeThe short video titled ” was also one of the most popular examples.
However, in different cases, the experiences of other users and movie industry figures have not been uniformly positive. Many content creators believe that Sora can be used in complex scenes. Inability to perform consistent physics simulations and given failure to fully comply with instructions He stated that they encountered such problems. For example, in a video shared by a user, the legs of people walking are moving in the opposite direction and anatomically unreal results were observed.
Among Sora’s biggest rivals RunwayLuma, Hailuo, Kling and Tencent’s Hunyuan models. In particular, HailuoAI is stated to be more effective and easy to use by users. Unlike OpenAI, while many competing models offer free usage tiers, Sora’s lowest paid plan is $20 per month, while its unlimited usage option comes with a hefty price tag of $200.
Additionally, OpenAI’s content restrictions have also been the focus of criticism. The ban on violent and obscene content, as well as even “cartoon-style” elements, limits the creative potential of the model for some users.
It’s hard to get Hollywood’s attention
Sora’s “sketchy” start could also jeopardize OpenAI’s plans to collaborate with Hollywood studios. From your competitors Runwaycurrently Lionsgate has signed a special agreement with. With this agreement, the firm will provide Lionsgate with AI models specially trained on a catalog of more than 20,000 movies and TV shows.
In OpenAI’s blog statement, the model is still There are many areas for improvement is accepted: “Sora may have difficulty simulating the physics of complex scenes and may not understand certain cause-effect relationships. “The model may also confuse spatial details included in a prompt, such as right-left distinction, or fail to accurately convey events that develop over time, such as certain camera trajectories.”
Sora’s current situation is difficult to interpret. Although it can sometimes produce very realistic results, it can also produce videos that are ridiculous and devoid of physics or anatomy. This consistency issue means that Sora is behind in this race for now.
This news our mobile application Download using
You can read it whenever you want (even offline):
