Microsoft Suspends Bing Image Creator Template Update After Quality Criticism | Borisdayma dalle mini github | Best ai medical diagnosis app | Ai image generator from image free online | Turtles AI

Microsoft has temporarily withdrawn the new PR16 model of DALL-E 3 for Bing Image Creator after numerous complaints about image quality. Users report inferior results compared to the previous version, prompting the company to temporarily reinstate the old template.

Key points:

• New model failed: The PR16 model promised speed and quality, but disappointed users.
• Widespread complaints: Users describe less realistic images and “cartoony” quality.
• Temporary Reset: Microsoft reverts to PR13 model to address reported problems.
• Challenges for AI companies: Case highlights difficulties in ensuring effective improvements in the field.

Microsoft finds itself at the center of a heated debate after an update to the AI model behind Bing Image Creator sparked a wave of criticism. The tech giant had announced that the new model, identified by code PR16 and based on the latest version of DALL-E 3 developed by OpenAI, would offer a faster experience and higher quality of generated images. However, after the release, users on platforms such as Reddit and X expressed dissatisfaction, pointing out that the PR16 model produces less realistic images with poor detail and an artificial, flat look. Some described the switch as a significant regression, complaining about the loss of the quality and precision they had enjoyed in previous versions.

The negative reactions forced Microsoft to act quickly. Jordi Ribas, head of the research division, confirmed that the team was able to reproduce some of the problems reported by users. In response, Microsoft decided to temporarily revert to the previous model, PR13, to ensure a better experience, at least until the new system’s flaws are fixed. However, Ribas warned that the restoration process will be gradual and take weeks to complete, leaving some users still dissatisfied in the short term.

This affair is not an isolated case in the generative AI landscape. Other companies have also encountered difficulties in balancing innovation and public acceptance. A recent example is Google, which in February suspended the image generation function of its AI chatbot Gemini because of errors in historical representations. The issues encountered with PR16 highlight how complex it is to translate progress made in development into improvements perceptible to end users. Although Microsoft’s internal benchmarks had indicated a slight improvement over the previous version, this was not reflected in the user experience.

The main challenge for the tech giants, then, seems to be to align technical benchmarks with public expectations, ensuring that each new technological breakthrough is a real advance from the user’s perspective as well.