OpenAI has added a Sketch feature to ChatGPT, which allows users to create images based on their own drawings. Now, it's enough to draw a simple diagram directly in the chat, and then give the AI a text description of how exactly it should be detailed. A simple sketch of a cat, made with a mouse, easily turned into a realistic photo following instructions. This became possible thanks to the latest update of ChatGPT Images 2.5.
This capability breaks down the barrier between an idea and its visualization. If previously it was necessary to precisely formulate a request or use complex tools, now it is enough to draw something resembling the desired object. The function is activated by entering @Sketch in the chat, after which a drawing window appears. Testing has shown that even a careless sketch made with a computer mouse is successfully transformed into a detailed image in accordance with text instructions. This makes image generation more intuitive and accessible for those who do not have design skills or complex prompts.
The new tool opens doors for more personalized creativity. Artists, designers, marketers, and enthusiasts alike get the opportunity to bring their ideas to visual format faster. It will be possible to experiment with concepts, create illustrations for personal projects, or even product prototypes, simply by sketching them by hand. This is no longer just generation based on a text description, but a dialogue where a visual sketch becomes the starting point.
This is another step towards artificial intelligence becoming not just a tool, but a full-fledged creative partner that understands even the vaguest intentions. In the future, such an approach can significantly accelerate the process of prototyping and visualizing ideas, making creativity more accessible to everyone.