Yes, really! Paint is downright useful now.

Like many, I’ve been using generative AI (genAI) systems for about two years. Nothing has quite triggered my imagination like Microsoft Paint’s Cocreator capability.

The concept is straightforward: You can doodle using Paint, describe what you want, and genAI will create an image to your specifications.

I can’t shake the feeling that this way of working is going to shape the future of collaboration in enterprises. We can see this actively unfolding, coming to more apps you use thanks to AI being included in more basic licenses. Microsoft is enabling Copilot Chat in all enterprise accounts, Google Workspace is including Gemini in its business, and enterprise plans of Zoom include Zoom AI Companion in paid licenses. Embedded cocreating AI is increasingly available, and Paint provides a preview of what interacting with these systems may look like.

How Paint Cocreator workflows change modern work

As with all genAI systems, showing is better than telling. Let’s walk through the workflow process with visuals on how AI-enhanced Paint with Cocreator changes today’s modern work, workforce dynamics, and productivity.

Below is a very poor, very quick rendering of a butterfly using the marker tool:

 

The panel to the right is Cocreator and its suggestion:

 

Taking that suggestion, I can place it in the canvas and directly annotate. For example, let’s push the wings to be a bit bluer by painting over them with a marker:

 

Here, I took the generated image and doodled over it with the pen. The colors are much closer to what I want. Here’s what it looks like, a bit more realistic, turning off the stylization:

 

But the scene is kind of boring — butterflies deserve flowers. Let’s put it in a field of them:

 

Scribbling broadly over the background allows me to generate something a lot more detailed:

 

Further markup allows me to create the broad shape of my vision, but I’m still limited by my own inability to be effective with a trackpad. So I’m cutting Cocreator loose and telling it to run wild creatively with the creativity slider:

 

Here’s the end result:

 

I started with this:

 

All I used was a combination of text prompts and Paint’s marker tool to adjust.

This is the new Microsoft Paint: the universal design app on every PC. The system isn’t without some major flaws — like resolution, inability to generate accurate living creatures, some truly upsetting accidental generations, and all generations suffer from the “AI bloom” issue. But Paint is fun to use again. I feel like I’m back in my elementary school computer lab, realizing that on the computer, I could draw anything I wanted.

Why does this matter for businesses?

In embedding genAI into established application interfaces, Microsoft has proven the AI value proposition in ways many organizations still struggle to do.

It provides the ability to go from vague draft to fleshed-out concept in minutes — massively accelerating cycle time, allowing me to directly tweak specific generation components to better meet what I’m trying to accomplish directly in the GUI. And the outputs are transferable. Try it as soon as you can.

Some of my favorite generations so far:

 

 

 

You’ll note that the model doesn’t generate text well yet.

Have questions? Schedule an inquiry with me.

(Access to Paint Cocreator was provided courtesy of Microsoft for testing.)