OpenAI has announced improvements to DALL-E, its AI image generator that can now be integrated with ChatGPT

OpenAI, a company dedicated to developing AI-based technologies Artificial Intelligence (AI)advertisement Improvements to his imaging tool, DALL-E, which has already reached its third edition. The most important new thing is that users will be able to use the company’s chatbot, ChatGPT, to get more complex results from simpler commands.

Integrating DALL-E 3 with ChatGPT means that those using the image generator will be able to rely on chatbot assistance in Simplifying, in this way, the so-called “claims”These are the text commands that the system relies on to generate results.

The difference from the previous version of the application is that to obtain more detailed images, it was necessary to include specifications specific to the design field, a type of knowledge that not all users possess. from now on, Less detailed instructions can be given to the chatbot to create an action, thus improving it.

This way, the user will be able to write a short text with their request for the image, and ChatGPT will “fine-tune” the technical details.

What is DALL-E, the OpenAI image generator

The name of this system comes from the combination of Pixar characters Wall-E and artist Salvador Dali. Launched in 2022, DALL-E is powered by an algorithm trained on millions of labeled images taken from the web and other sources.

From all those reference photos, What the application does is determine how the image will be displayed based on the user’s request.

For example, the DALL-E 3 system generated an image from the following request: “Detailed oil painting of an old sea captain, steering his ship through a storm. Salt water splashes over his damaged face.” The determination is reflected in his “eyes. Above, malevolent clouds hover and stern waves threaten to sink the ship as seagulls dive and swirl in the chaotic landscape. Thunder and lights sail in the distance, illuminating the scene with an eerie green glow.”

he As a result of that claimThe company published the following on its official website:

Once the result is ready, the user can continue giving instructions to the chatbot to improve that image. “You won’t have to bother with really long prompts,” said Aditya Ramesh, principal researcher and leader of the Dall-E team, adding: “Instead, you can simply interact with ChatGPT as if you were talking to a co-worker.” .

OpenAI’s efforts to improve its AI-based image generator comes in a context where its competitors, including alternatives like Midjourney and Stable Diffusion, have demonstrated remarkable abilities to deliver high-quality results.

DALL-E 3 is now available through ChatGPT Plus, the paid version of the chatbot.

Bid on the copyright of the works used

Since its launch, DALL-E has been trained using images produced by artists from around the world, and controversy has arisen following complaints from authors who did not consent to the company’s use of their work.

That’s why OpenAI announced it as well It will provide artists with the ability to remove their work from future training. Additionally, DALL-E 3 will also prevent users from trying to create artwork in the style of a known artist by detecting when it is included in a question.

OpenAI has too Barriers To prevent users from creating pornographic or violent artwork or images featuring public figures.