After the success of Chatgpt, OpenAI continues to surprise with its releases, this time it’s Sora‘s turn, the text-to-video AI tool. This model is capable of creating videos that look like something out of a fictional movie with just a descriptive paragraph that we call “Prompt”.
Currently, it’s not available for the general public, but for some artists, and creative users who are testing its potential.
With just a few lines of text, Sora can interpret exactly what we want for the video, similar to when you’re told a story and imagine the scene in your mind. You can describe the characters, emotions, environments, perspectives, weather, actions, and more to generate detailed videos. Here are some videos released by the OpenAI X account:
Prompt: “A stylish woman walks down a Tokyo street filled with warm glowing neon and animated city signage. she wears a black leather jacket, a long red dress, and black boots, and carries a black purse. she wears sunglasses and red lipstick. she walks confidently and casually.… pic.twitter.com/cjIdgYFaWq
— OpenAI (@OpenAI) February 15, 2024
Some inconsistencies are still observable in the videos, and the model still shows flaws in interpreting the texts. Currently, all videos generated with Sora contain a watermark, although OpenAI admits that it is easily removable.
There is no doubt that Sora is capable of creating impressive videos, however, the OpenAI red team is in the process of making this tool as secure as possible to avoid privacy and intellectual violation, or the generation of unsafe visual content.
If you’re keen on exploring the exciting tools that can be developed with AI, we invite you to take a look at two of our most successful projects: Writerly and Ekom.
Leave a Reply