ChatGPT Sketch turns doodles into AI images — SkimNews
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- OpenAI announced ChatGPT Images 2.5 on Tuesday alongside a new Sketch feature that lets users draw inside the chat to guide image generation.
- The Sketch tool is activated by typing @Sketch in the chat box, opening a drawing window; a tester turned a mouse-drawn cat doodle into a realistic photo via text instructions.
- Users can now leave inline comments directly on parts of an image to request targeted edits, such as changing a cat's eyes from blue to green.
- ChatGPT Images 2.5 delivers "more natural lighting and richer textures" and follows editing instructions more reliably across multiple turns, per OpenAI.
- Image generation latency dropped by up to 50% compared with Images 2.0.
- Images 2.5 is available now to ChatGPT, ChatGPT Work, and Codex users on desktop, mobile, and web.
Why it matters: By letting a doodle substitute for descriptive prose, OpenAI removes a key bottleneck for users who can't articulate what they want — widening the audience for AI image generation. The 50% latency cut shortens iteration loops for the ChatGPT, ChatGPT Work, and Codex user bases, making visual workflows cheaper in time even without a quality-vs-cost trade-off.
Ask SkimNews
