Meta Launches Imagine with Meta, a Standalone AI-Powered Image Generator

Meta has launched Imagine with Meta, a standalone AI-powered image generator that allows users to create high-resolution images from text prompts. The tool is powered by Meta’s existing Emu image generation model and is free to use for users in the US, generating four images per prompt. The company is also adding invisible watermarking to increase transparency and traceability of AI-generated images. The Imagine tool was previously available only through individual and group chats in Meta’s social network platforms, but now it is freely accessible on the web for anyone in the US to try.

Meta’s Imagine with Meta is similar to OpenAI’s DALL-E, Midjourney, and Stable Diffusion, which also use natural language to generate images. The tool is designed to be more playful and back-and-forth, but it can now be used to create free images on the web. Meta’s Emu model can fashion high-quality, photorealistic images within seconds, and the image generator can produce realistic images within a few seconds. 

Meta is adding more safeguard elements, like invisible watermarking, to track generative AI content and ensure greater transparency. The company is also continually testing its AI systems to mitigate “risky outputs.” Meta is making its chat personas available to more users, which will let more people engage with AI versions of Kendall Jenner, Tom Brady, and more. The company is also adding real-time discovery capacity, via Bing, to more of these personas. 

Meta’s Imagine with Meta is a significant step forward in AI-generated images, but it also raises concerns about transparency and traceability. Recently, China’s Cyberspace Administration issued regulations requiring that generative AI vendors mark generated content, including text and image generators, without affecting user usage. In recent US Senate committee hearings, Senator Kyrsten Sinema emphasized the need for transparency in generative AI, including by using watermarks.

Meta Launches Imagine with Meta, a Standalone AI-Powered Image Generator, allowing users to create high-resolution images from text prompts. The tool is powered by Meta’s existing Emu image generation model and is free to use for users in the US, generating four images per prompt. The company is also adding invisible watermarking to increase transparency and traceability of AI-generated images.