What happened

OpenAI announced ChatGPT Images 2.5 on Tuesday, introducing a new feature called Sketch that allows users to draw a doodle directly inside ChatGPT and then instruct the AI to generate an image based on that sketch.

The Sketch feature is designed to provide a more intuitive way for users to communicate their visual ideas to ChatGPT, bridging the gap between rough drawings and polished AI-generated images.

Why it matters

This feature could make AI image generation more accessible to people who struggle to describe images in words, as a simple doodle can serve as a visual prompt.

By allowing users to iterate on their sketches with text instructions, it may enhance creative workflows, enabling faster prototyping of visual concepts.

Key facts

OpenAI announced ChatGPT Images 2.5 on Tuesday.

The new feature is called Sketch.

Sketch allows users to draw inside ChatGPT and then tell ChatGPT how to turn the drawing into an image.

What to watch next

It remains to be seen how well Sketch handles complex or abstract doodles, and whether it will be integrated into other OpenAI products.

The rollout of ChatGPT Images 2.5 and Sketch may signal a broader trend toward multimodal input methods in AI tools.

Sources