Rewriting a full AI image prompt just to fix one element is one of the more tedious parts of working with generative image tools. Google's new Pics app takes a different approach.
Announced at I/O 2026, Pics is a standalone AI image creation and editing app inside Google Workspace. The core mechanic: instead of crafting a new prompt when you want to change something specific, you click directly on that element and describe only what you want different. Google says the surrounding image stays intact.
Adobe Firefly and Canva both handle targeted edits through a technique called inpainting - you draw a selection mask over the area you want changed, then write a prompt for just that region. Pics appears to remove the masking step, collapsing the workflow into a single click-and-describe interaction.
The app is launching inside Google Workspace, targeting business users rather than the broader consumer market. No separate pricing was announced; access will likely follow standard Workspace plan tiers.
The real test is whether Google's underlying Imagen model can reliably edit one area of an image without changing what's around it. That's been the persistent hard problem for every AI image editor: change the background color and a face shifts; edit a shirt and the lighting moves. Controlled demos don't typically surface these failure cases. They appear when actual users give imprecise inputs, which is most of the time.
If Pics holds up in practice, the click-to-edit workflow is a genuine improvement over current AI image tools. If it doesn't, it joins a long list of AI image features that work better on stage than in daily use.