DALL-E is a powerful AI-based image generator that works similarly to other neural networks, generating results (images) based on text description. It will now be integrated into Bing Chat and Edge, extending its functionality beyond bot and text chat.

what is known

By entering an image description, providing additional context such as location or activity, and choosing an art style, Image Creator will create an image from your own imagination.
– says Yusuf Mehdi, head of consumer marketing at Microsoft.

According to the statement, Image Creator will appear as a new icon in the Edge sidebar. It can be enabled in Settings in the section of the same name. In the latest version of the company’s browser, a new button has already appeared next to the address bar to search for Bing Chat. It lets you open a modal window by clicking it, or a temporary window that will hide itself by hovering the cursor.

This will make Edge “the first and only browser with an integrated renderer based on artificial intelligence”. Interestingly, users who still do not have access to Bing will be able to use Image Creator. For now, you can do this by following the link.

Microsoft shared a few examples of how artificial intelligence will work.

• In one of them, a user asks Bing to “give me an image of an astronaut walking through a galaxy of sunflowers.” The system outputs 4 images in response.
• In another screenshot from Microsoft, a user asks Bing to create a living room image for inspiration, and we see how that works right in the sidebar. Image generators provide four more examples. The author of the request chooses one of these pictures and publishes it in his feed on social networks.

Microsoft will be careful and gradual in providing access to the service to avoid negative experiences with Bing, where journalists and other users use the chat to generate awkward, sometimes awkward responses. The Image Maker will also be available in the “creative” mode, where initially only artificial intelligence creates more original and creative images. The other two modes, “balanced” and “precise”, will reportedly be introduced later.