I tested the new ChatGPT Images - it's a stunning improvement, and enormously fun (www.zdnet.com)

OpenAI has unveiled a significant update to ChatGPT's image generation capabilities, now branded as "the new ChatGPT Images" or GPT Image 1.5. This release follows the recent rollout of GPT-5.2 and introduces advanced features including enhanced recontextualization of images, allowing users to edit and composite elements into existing photos effectively. Notably, the update is accessible to all ChatGPT tiers, including free users, marking a step towards democratizing powerful image generation tools. This improvement is vital for the AI/ML community as it reflects OpenAI's commitment to enhancing user experience and pushing the boundaries of generative AI. User testing revealed that the AI successfully modified detailed aspects of images upon user prompts, although some unexpected changes occurred in the output, showcasing both the strengths and limitations of the technology. The new model facilitates creative fun—a testament to its engaging design—while highlighting areas for enhancement, such as maintaining original image dimensions and improving visual consistency. Overall, this update is poised to broaden the applications of AI in creative domains, inviting users to explore and experiment with image generation like never before.