The Inktelligence - December 16, 2024

The end of the year has been shaping up to be a busy one in terms of new AI releases. Let's dive into the most significant developments that are shaping the future of AI.

Open AI finally reveals Sora to the public

OpenAI had generated a lot of interest in Sora with their sneak peaks in the past but this is the first time we are able to get some hands on experience with it. Sora, their text-to-video AI model has been targeted towards the creative industries, from advertising to filmmaking.

The introduction was not without hiccups as many, including myself, were not able to login to the platform due to the load on the system.

When I finally had access, I went on to create some videos.

The UI is familiar, with many AI video generators using the same format of a screen displaying videos that are generated by the community.

You just type your prompt and select the options from:

Presets: None, Balloon World, Stop Motion, Archival, Film Noir, Cardboard & Papercraft

Aspect ratio: 16:9, 1:1, 9:16

Resolution: 480p (default), 720p (4x slower), 1080p (8x slower but this was not available to me at the $20/month ChatGPT subscription)

Video length: 5, 10, 15, 20s (at my subscription level, only 5 and 10s are available)

Number of videos: 1, 2, 4 (this is how many videos you can generate at one time)

There is another additional feature called the “Storyboard”. This is like an editing panel which will let you add scenes along the timeline. You can create text-to-video, image-to-video or video-to-video.

I found that more descriptive prompts tend to work better. So I took some help from ChatGPT to come up with a prompt describing a scene from the NYC subway. Here’s the prompt: