Meta is bringing a far more capable kind of image-making to Instagram and WhatsApp, turning its social apps into places where users can create, remix and edit visuals simply by describing what they want.

The new system, called Muse Image, is Meta’s first image-generation model from its Superintelligence Labs division. It is arriving through Meta AI, initially in a limited set of countries, and will eventually spread across Facebook, Messenger and Meta’s advertising tools too.

For most people, the immediate change will feel less like using a separate AI product and more like discovering a new creative layer inside apps they already open dozens of times a day. On Instagram, Muse Image powers more than 30 AI effects for Stories. On WhatsApp, people can generate images in direct conversations with Meta AI. Rather than starting with a blank prompt box, users can also work from an existing photo, add several pictures, or sketch directly over an image to point out the changes they want.

That matters because image AI has largely lived in dedicated tools such as ChatGPT, Gemini, Midjourney and Adobe Firefly. Meta’s advantage is distribution. Instagram and WhatsApp reportedly each reach more than 3 billion users globally, meaning generative image features could quickly become part of the ordinary rhythm of messaging, posting and reacting to friends.

A user might take a rough holiday photo and ask for a brighter sky, cleaner background or a more dramatic color treatment. Someone preparing a birthday post could turn a vague idea into a custom invite, then send it straight into a group chat or add it to a Story. Meta says Muse Image can understand conversational prompts, combine multiple photos and accept markups such as circles, annotations and sketches that indicate exactly where an edit should occur.

It is an attempt to make the technology feel less technical. Previous generations of AI image tools often rewarded people who knew how to write unusually detailed prompts. Meta is betting that its audience would rather say, “make this look like a vintage magazine cover,” circle an unwanted object, or select from suggested ideas, then move on with their day.

There is an obvious social-media logic to that approach. Instagram is already built around the impulse to make a moment look a little more polished, playful or imaginative before sharing it. AI expands the range of what can be changed. The line between a filter, an edit and a fully generated scene is becoming increasingly thin.

Muse Image also arrives at a sensitive time for online authenticity. AI-generated imagery can be delightful when it is clearly creative – a surreal birthday card, a stylised profile image or an impossible visual joke. But the same tools can complicate how people assess images that appear realistic, particularly when they are presented without context.

Meta has previously said it applies “Imagined with AI” labels to photorealistic images created using its own Meta AI tools. It has also worked with industry partners on technical standards designed to detect AI-generated imagery from other providers, using visible markers, invisible watermarks and metadata where available. Whether those signals remain clear and resilient after screenshots, reposts and edits will matter just as much as the quality of Muse Image itself.

For creators, the rollout is likely to be both tempting and disruptive. It lowers the time and skill required to make bespoke visuals, which could be useful for small businesses, community pages and individuals who want to give a post more personality without opening a design app. Meta says the tool is free for everyday creation, while heavier use is available through its subscription plans.

At the same time, easier creation means a platform already crowded with visuals could become even more synthetic. The challenge will not just be making images. It will be making images that still feel distinctive, trustworthy and worth someone’s attention. A polished AI-generated picture can stop a scroll, but it cannot automatically supply a creator’s perspective, humour or credibility.

Meta is preparing to take the technology beyond everyday posts. The company says Muse Image will reach advertisers and agencies through Advantage+ creative in the coming weeks. That could make campaign asset production quicker and cheaper, but it also raises familiar concerns around creative sameness, brand safety and disclosure when commercial visuals have been materially generated or altered.

For now, the biggest takeaway is straightforward: Meta is no longer treating generative AI as a separate chatbot feature sitting at the edge of its apps. It is putting it directly into the social and messaging experiences where people already communicate visually.

The practical question is not whether Instagram and WhatsApp users will encounter AI-made images. They will. The more interesting question is whether Muse Image becomes another disposable effect in the Stories tray, or a genuinely useful creative tool that gives people more control over how they express themselves. Meta has the audience to make either outcome enormous.