Designer Input

Welcome back to Designer Input,

This Week’s Topics:

• New Update from OpenAI: Dall-E 3

• AI is everywhere now: Microsoft Copilot

• 3 AI Tools for Designer2.0’s

• AI-Guide of the Week: SDXL Artist Style Studies

• Stories Worth Reading: Bard can now connect to your Google apps and services.

Dall-E 3

OpenAI has announced the third iteration of its generative AI visual art platform, DALL-E 3, which now integrates with ChatGPT, to generate prompts for creating visual content.

DALL-E 3 has improved context understanding and offers more safety features, including:

• DALL-E 3 won't mimic the art styles of living artists.

• Artists can opt their images out of training DALL-E's future models.

• DALL-E 3 will decline requests for public figures by name.

What excites me most about DALL-E 3 is its skill at understanding detailed prompts and getting the small details right. Usually text-to-image generative models, require prompts with special keyword tricks rather than natural language.

Of course, we have to try it hands-on to fully judge and see if the examples are cherry-picked to look ideal.

Initially, OpenAI will roll out DALL-E 3 to ChatGPT Plus and ChatGPT Enterprise users, with releases to research labs and API service users to follow. But it's unclear when a free public version of DALL-E 3 will be available.

P.S. Microsoft also announced that Bing chat will soon include access to the DALL-E 3 image generator, which will allow users to create images right in the chat.

AI is everywhere now. LITERALLY.

Announcing Microsoft Copilot, your everyday AI companion

Microsoft introduced Copilot, bringing AI-powered capabilities to Windows 11.

The new “everyday AI companion”, now called Microsoft Copilot in Windows, will be “seamlessly available across all the apps and experiences you use most”, including the Windows 11 desktop, Microsoft 365, Outlook, the Edge browser, Paint and more.

Paint is getting text-to-image generator “Cocreator” as well as Layers -yes, layers finally- and Background Removal feature.

The Copilot will be able to pull data from your calendar, email, documents, and more to help you quickly draft texts and emails, and it’ll make suggestions in real time in apps like Outlook.

The Copilot experience will be available on Windows 11, starting from September 26th.

 3 AI Tools for Designer2.0’s 

1) AutoDraw: It pairs machine learning with drawings from talented artists to help you draw stuff fast. (Free)

2) Unscreen: Remove the background of and video, automatically, online. No greenscreen. (Freemium)

3) AI Comic Factory: Generate a custom comic book from a single text input. (Free)

 AI-Guide of the Week (SDXL Artist Style Studies) 

The latest Stable Diffusion model, SDXL, can recognize the unique styles of thousands of artists. The images in this study were generated with the same settings and parameters, except for the name of the artists.

SDXL Artist Style Database consists of 4387 different artist styles. There are two sets of generated images with the following prompts:

• “a portrait of a character in a scenic environment by [artist]”

• “a building in a stunning landscape by [artist]”

Stories Worth Reading - Our Picks for You

• Microsoft AI Researchers Accidentally Expose 38 Terabytes of Confidential Data.

• New text-to-image model, Würstchen. It can generate images much faster than Stable Diffusion XL, while using a lot less memory!

• Bard can now connect to your Google apps and services.

• Generative Image Dynamics. Click and drag a point on the image to automatically transform it into seamless looping videos.

Thank you for reading and see you next time!

Cheers,

Do you have something in mind? Feel free to hit reply to this email. 👇🏻

I’m more than happy to hear your ideas, suggestions, or questions.

Want to advertise in the Designer Input? Check here.