DALL-E and Sora: OpenAI shows new AI developments

The AI ​​company OpenAI, best known for ChatGPT, demonstrated a few new developments in its image and video AIs this week. The image creation tool called DALL-E has been expanded to include the option for individual, text-based post-processing of generated images. The video creation tool called Sora, which is not yet available to the general public, has been tested by selected people. On the one hand, they have used it to create scenes and videos that convey extremely creative ideas. On the other hand, the videos can also be compared to the crazy things you dream up when you're half asleep.

Editing options using the image AI “DALL-E” (left) and various results of the video AI “Sora” (right) are currently being demonstrated. Source: OpenAI / YouTube
Editing options using the image AI “DALL-E” (left) and various results of the video AI “Sora” (right) are currently being demonstrated. Source: OpenAI / YouTube

OpenAI DALL-E – Created images can be edited multimodally

In an example video for the new graphics post-processing functions, OpenAI initially shows image creation with DALL-E. The AI ​​is asked to create a poodle that says “Happy Birthday” with a sign. The AI ​​then creates this  requested picture and decorates it with other objects that can be associated with a birthday party.

Then one of the two created images is selected, a tool for editing is clicked on and two places on the dog's ears are marked directly on the image. By entering text, the AI ​​is asked to add loops there, which is implemented after some processing time. This means you don't have to create a completely new image if only one or two improving details are missing from the result.

OpenAI Sora – Several example videos for a variety of applications

Auf dem OpenAI YouTube channel Various videos have been and are currently being uploaded that demonstrate the possibilities of using Sora should show. In addition to a music video by August Kamp that combines dream worlds with patterns and effects, there is a fantasy documentary called “Beyond Our Reality” by Don Allen Stevenson that shows various animal chimeras.

A video by Josephine Miller shows people wearing glass clothing or made of glass bodies underwater. And finally there is a video by Paul Trillo in which gold turns from an earth element into a record and flies into the world. All videos impressively show how far AI has already come. However, in every video there are also elements, movements, camera movements, glitches and other individual factors that make the material seem unreal and sometimes unpleasant.

Here are the videos:

My tips & tricks about technology & Apple

Did you like the article and did the instructions on the blog help you? Then I would be happy if you the blog via a Steady Membership would support.

Post a comment

Your e-mail address will not be published. Required fields are marked with * marked

In the Sir Apfelot Blog you will find advice, instructions and reviews on Apple products such as the iPhone, iPad, Apple Watch, AirPods, iMac, Mac Pro, Mac Mini and Mac Studio.