OpenAI has announced a significant update in ChatGPT's image generation, reinforcing its commitment to turning the platform into a true digital creative studio. This new version stands out for its accuracy in following instructions, allowing for edits that are more faithful to user requests while maintaining details such as facial likeness, lighting, composition, and color tone across different modifications. The generation speed has also improved notably, now being up to four times faster, which facilitates a much smoother and more practical experience. Additionally, a sidebar with predefined styles and trending ideas has been incorporated, aimed at users without technical experience seeking visual inspiration. OpenAI's goal is to position ChatGPT as a versatile tool for daily image editing, creative transformations, and real-world applications, directly competing with Google's Nano Banana technology, which has set trends in recent months. This update follows the recent integration of Photoshop Express and other Adobe applications, expanding editing possibilities within ChatGPT. The new model, available from today in both ChatGPT and through the API under the name GPT-Image-1.5, allows users to combine multiple images, add elements, and transform photographs into creative compositions, such as turning an image of two engineers into a retro movie poster. Text rendering has also been optimized, achieving the insertion of lengthy texts in complex formats, like newspapers, with precision that surpasses the competition. This launch coincides with the arrival of GPT-5.2, OpenAI's new AI model, which aims to regain ground against Google's Gemini 3 Pro, especially in multimodal and reasoning capabilities. In recent weeks, competition between the two companies has intensified, and OpenAI is responding with innovations that strengthen its position in the sector. (Source: elespanol.com)