ChatGPT Sketch converts rough sketches into detailed AI images
openai
| Source: The Verge | Original article
OpenAI's ChatGPT Images 2.5 adds a Sketch feature that turns user doodles drawn inside ChatGPT into detailed AI-generated images.
OpenAI rolled out ChatGPT Images 2.5 on Tuesday, adding a feature called **Sketch** that lets users draw a doodle directly inside ChatGPT and use that sketch as part of the prompt for image generation. The tool interprets even crude line drawings and expands them into detailed, polished images, effectively turning a “bad drawing” into a finished visual output.
The addition matters because it lowers the skill barrier for creating high‑quality AI‑generated art. By accepting hand‑drawn sketches as visual cues, Sketch expands the ways users can communicate intent, making the platform more accessible to non‑artists and speeding up the creative workflow. OpenAI highlighted that its image service already powers more than three billion images each week, and the new capability could further boost engagement by simplifying the prompt‑crafting process.
As we reported on 9 September, the Images 2.5 update also cut generation latency by up to 50 percent compared with Images 2.0 and introduced a Sketch tool for drawing in ChatGPT. The latest release adds two companion models to the API, signalling that OpenAI is extending the feature set beyond the consumer interface.
What to watch next includes how quickly developers adopt the new API models, whether third‑party tools integrate Sketch for richer visual creation, and how competitors respond with similar sketch‑to‑image capabilities. User feedback on the quality of the generated results will also shape the next iteration of OpenAI’s image‑generation roadmap.
Sources
Back to AIPULSEN