Meta has updated its AI image generation tool for advertisers. This upgrade aims to help marketers create more diverse ad variations based on current content trends that are performing well on Facebook and Instagram.

The new process uses the Muse Image model developed by Meta's Superintelligence Labs AI research team. The updated AI image generation tool will expand marketers' ability to restyle existing ad creatives, while also supporting the direct generation of static images from video assets.

Muse image for ads

Meta explained: "Muse Image has agentic visual reasoning and self-improvement capabilities, which means it can understand complex creative briefs like a designer, rather than just recognizing isolated keywords. Muse Image brings native reasoning into the creative process, allowing it to adjust elements, switch styles, and generate variations based on the advertiser's assets, thereby producing high-quality, brand-aligned ad variations with fewer iterations."

In addition to leveraging Meta's most advanced AI systems, the tool is also built on Meta's systematic understanding of promotional activities, which enables it to generate content that better aligns with the current response trends of audiences across its apps.

Meta said that advertisers who previously used its AI tools to generate multiple promotional variations have achieved significant results. Early testers of Muse Image reported that the system can produce higher-quality creative content, including more realistic photorealistic effects and more reliable preservation of product integrity.

Muse Image will also bring a greater expansion in generation capacity.

Meta said: "Meta AI can now reimagine consumers' rooms, a capability powered by Muse Image's photorealistic image generation technology. When users upload photos of their spaces, Meta AI restyles them using real products from merchants' catalogs, showing how each item looks in the actual room. Users can compare different options, adjust the visuals through ongoing conversation, or click through directly to the brand's website to make a purchase."

This expanded visual exploration process is Meta's next step in its AI advertising initiatives, aimed at providing users with new ways to interact with and engage with promotional content.

Meta has invested heavily in AI infrastructure and is now seeking to recoup these costs through a series of AI innovations.

Muse Image is Meta's latest project in this direction, with its Superintelligence team working to build better tools for a broader range of customers.

For this reason, the success of these models is of great importance to Meta. This also means that exploring its generative AI options may be worth trying.

Meta revealed that more than 8 million advertisers are currently using at least one of its generative AI ad creative tools.