ToolsSeptember 8, 2026via The Verge AI
ChatGPT Sketch turns your bad drawings into detailed AI images
Why it matters
OpenAI ships a new interaction model for image generation (Sketch) as part of ChatGPT Images 2.5, lowering the bar for users who can't write detailed prompts. Practitioners building image workflows should evaluate whether sketch-based input changes their adoption calculus.
Key signals
- ChatGPT Images 2.5 announced Tuesday
- New Sketch feature lets users draw inside ChatGPT and refine via prompt
- Activation: type @Sketch into chat box to open drawing window
- Successfully converted hand-drawn doodle (mouse-drawn cat) to photorealistic image
- New comment-on-images feature also shipped
- New Sketch feature activated via @Sketch command in chat
- Converts hand-drawn doodles into detailed images via text instruction
- Tested on cat drawing—mouse-drawn doodle converted to realistic photo
- Inline image commenting feature also added
- Available inside ChatGPT interface (user-facing product feature)
The hook
ChatGPT Sketch turns doodles into photorealistic images—the new UX for image generation just got more accessible.
I used ChatGPT and its Sketch tool to make this AI-generated image of a cat.
OpenAI announced ChatGPT Images 2.5 on Tuesday and is adding a new way to tell ChatGPT what you want it to make an image of: by drawing a doodle. With a new feature called Sketch, you can just draw something right inside …