Last week, OpenAI announced a new AI model called Sora that can generate video clips of up to 60 seconds based on text. The company also added a number of generated video clips to the news, which look very good and are a big step forward compared to the tests seen so far.
OpenAI reports that Sora is currently being made available to test teams to assess critical encounters for damage or risk. These teams include experts in areas such as misinformation, hate speech and bias.
In addition to this testing, Sora is also said to follow the safety measures in place for DALL·E 3. The company adds that it is working on tools to help identify whether a video was created by Sora.
While other apps like Pika and Stability AI have beaten OpenAI when it comes to AI video generation, there are a few things that make Sora stand out. Sora can create videos of up to 60 seconds, while competitors can only manage around four seconds. Also, the sharpness, resolution and accuracy of the surrounding world is better with Sora.
There are currently over 35 examples on the OpenAI website. While the results are impressive, the model is far from perfect. The company admits that the model has weaknesses. It may struggle to accurately simulate the physics of a complex scene and may not understand specific cases of cause and effect. For example, a person may bite into a cookie, but the cookie may not have a bite mark afterward.
The model may also confuse spatial details, such as confusing left and right, and may struggle with accurate descriptions of events over time, such as following a particular camera trajectory.
Source: Android Authority