FILE PHOTO: OpenAI and ChatGPT logos are seen in this illustration taken, February 3, 2023. REUTERS/Dado Ruvic/Illustration/File Photo
During a live-streamed event on Tuesday, March 25, 2025, OpenAI revealed that ChatGPT will now be able to produce better images with coherent text. This is a major update that will allow ChatGPT to edit images and create visuals for work that include lengthy, readable text.
By making it simpler to alter photos in ChatGPT and provide graphics for projects that contain long, readable text, OpenAI may increase the chatbot’s attractiveness to both regular users and enterprises.
The model may be asked by users to produce a realistic picture of a personalized menu.
According to OpenAI, it will be able to follow more intricate human instructions about a photo’s arrangement.
The maker of ChatGPT includes state-of-the-art picture characteristics that may find application in both personal and professional contexts.
The San Francisco-based business demonstrated how ChatGPT customers would be able to improve photographs through a series of discussions with the chatbot during a live streamed event on Tuesday. For instance, a user may ask for a picture of a snail in a metropolis, then request that the backdrop be altered and a hat be added.
With the addition of image generators, ChatGPT’s features help it stay ahead of rival chatbots like Elon Musk’s xAI.
The business also claims that ChatGPT will be better equipped to produce pictures with logical language, which will let it produce diagrams, infographics, and logos for professional applications. For example, a user may ask the chatbot to create a realistic picture of a map or a personalized menu. According to the business, it will also be able to execute more complex user instructions on image composition.
OpenAI has made a greater effort to market ChatGPT as a multipurpose program that includes a voice assistant, video generator, and search engine. OpenAI is attempting to integrate more sophisticated picture characteristics with the most recent advancements, which may find use in a variety of personal and professional contexts. Additionally, it may help ChatGPT stay ahead of other chatbots, such as Elon Musk’s xAI, which have already included picture generators.
However, like other AI applications, ChatGPT could fabricate elements while creating photos, including adding text with fictitious national names to an image. The business stated in a blog post that consumers’ less thorough suggestions may result in these blunders. Among other things, OpenAI stated that their AI may struggle to produce small-sized writing and text in non-Latin alphabets.
According to a corporate blog post, the enhanced capability can also take up to a minute to generate photos. It takes longer since the visuals are more detailed, according to OpenAI CEO Sam Altman during the broadcast.
OpenAI’s GPT-4o model, which is accessible to both free and paying users, will make the additional features available on Tuesday. According to the corporation, software developers that utilize its application programming interface will be able to access the new functionalities over the course of the upcoming weeks.
We use cookies to ensure that we give you the best experience on our website. If you continue to use this site we will assume that you are happy with it.
