What happened? In October Microsoft AI presented its internal text-to-image conversion system, named MAI-Image-1. Well, as Microsoft AI CEO Mustafa Suleiman announced at X, the rendering model is now available Second pilot. Touted as a “huge improvement” over the previous system, you can start using the model right away – no long waiting lists, no experimental flags. Just open Copilot online or on your desktop and start generating.
- MAI-Image-1 is there designed for photorealistic resultsaddressing the historically dirty details, especially the hands, with much greater precision.
- Model supports creating text messages and editing imagesoffering refinement tools similar to those you’ll find in DALL-E.
- Users are already testing it publicly and First impressions are quite positive.
Why it matters: MAI-Image-1 marks an important moment for Microsoft because it’s the company’s first serious, fully proprietary image generator, meaning it no longer depends on third-party models like OpenAI’s DALL-E. This means Microsoft can now iterate faster, tailor the tool more to Copilot, and better serve regions where competing models (such as Google’s Gemini) have limited availability.
Early testers are already calling it a “huge improvement,” pointing to noticeably better photorealism, more natural lighting, and, most importantly, a major fix for one of the most obvious flaws in the art of artificial intelligence: weird fingers. Some developers who’ve tested the model say they’re impressed with how consistent and responsive it is, even praising it as “very good” for its detailed conceptual work.
It also shows where Copilot is headed: towards becoming a more complete creative platform than just a text assistant. Photo-realistic samples posted online, such as close-ups of hands in various environments or a realistic coffee cup with visible steam, suggest that Microsoft is closing the gap with established AI artistic tools. This raises the stakes in an increasingly crowded market where Meta, Google, Grok and OpenAI are competing to deliver best-in-class generative imaging. If Microsoft keeps up this pace, Copilot could become a space for creating images rather than just random prompts.
Why should I care? If you’re already using Copilot, your image experience just got better automatically, without the need to upgrade. Just upgrade to the new model and you’re good to go. Whether you’re creating mood boards, product concepts, thumbnails, illustrations, or just meme chaos, MAI-Image-1 is polished enough to make the process easier and more fun. And because it’s built right into Copilot, you don’t have to juggle external tools—it all lives where you’re already dialing in your routes.
OK, what’s next? MAI-Image-1 is now available in consumer Microsoft AI tools such as Copilot and Bing Image Creator, currently available in most major markets including the US, UK, Canada, Australia and other countries where Copilot supports text prompts. At the same time, Mustafa clarified that the model will “soon arrive” in the EU. To try it out, simply sign in to your Microsoft account, open Copilot (or Bing Image Creator) and enter a message to create or edit an image. Because the model is built into existing tools rather than a stand-alone site, it feels like a natural upgrade rather than a stand-alone learning application.
Looking ahead, Microsoft plans to expand the release further, including additional regions, languages, and platform integration. Early user feedback shows that the model benefits from photorealistic detail, refined lighting, and more natural anatomy, so expect Microsoft to improve editing workflows, add more style controls, and possibly open up model selection capabilities. In other words, this release isn’t the finish line, it’s the beginning of Microsoft’s claim to imaging, and each new iteration can unlock richer creative workflows in Copilot.
