One of the first companies you think of when it comes to artificial intelligence OpenAISince the day it introduced its chatbot ChatGPT, it has continuously fed itself with new features. In this context, ChatGPT, with innovations such as GPTs, plugins and DALL-E integration, has become an indispensable tool for artificial intelligence users.

But you know that when you create visuals with artificial intelligence find fault This is not a surprising situation. Especially when there are people in the picture. To edit them later, programs such as Photoshop are needed. In other words: it is difficult to get a perfect picture from artificial intelligence in one go.

Fortunately, ChatGPT makes it possible to edit images directly with the new update.

The post shared via OpenAI’s X account shows how the feature works. First of all, as usual, we provide a recipe to create our visual. When we click on one of the images, there is a button at the top right. brush icon welcomes us. When we press this, our cursor becomes one turns into a brush. We can also adjust the size from the top left.

We need to mask the parts we want to change using this brush. The text “Edit selection” will then appear in the text box. Then we rewrite what we want to change or add in those sections. In the example above bows added to the dog’s ears We see.

It is also possible to try out different styles.

When you get to the DALL-E page, you can try whatever you want in the text box. style examples It’s laying down. Examples include gothic, science fiction and synthwave.

Users who watched the first video complained that the masking editing method took a long time. Let’s hope this will be resolved with optimization efforts. At this stage, we would appreciate sharing the time in the sample video without cutting it.