Meta's Standalone AI Image Generator Launched

Like OpenAI's DALL-E, Midjourney, and Stable Diffusion, Imagine with Meta makes use of Meta's Emu image generation model to produce high-resolution images from text prompts.

In response to Google's Gemini artificial intelligence (AI) model launch, Facebook parent Meta is introducing Imagine with Meta. ( Image Source :Meta )

In response to Google's Gemini artificial intelligence (AI) model launch, Facebook parent Meta is introducing Imagine with Meta, a standalone generative AI experience on the web. This tool ill let users generate images by describing them in natural language. The company will roll out invisible watermarking to the new imagine with Meta AI experience in the coming weeks, in a bid to increase transparency and traceability of AI-generated images.

Like OpenAI's DALL-E, Midjourney, and Stable Diffusion, Imagine with Meta makes use of Meta's Emu image generation model to produce high-resolution images from text prompts. Currently, it's free for users in the US, generating four images per prompt, at least for now.

"We’ve enjoyed hearing from people about how they’re using imagine, Meta AI’s text-to-image generation feature, to make fun and creative content in chats. Today, we’re expanding access to imagine outside of chats, making it available in the US to start at imagine.meta.com. This standalone experience for creative hobbyists lets you create images with technology from Emu, our image foundation model. While our messaging experience is designed for more playful, back-and-forth interactions, you can now create free images on the web, too," the company said in a statement.

Meta said it is prioritising responsible development with a focus on safety across our products, understanding the crucial role of transparency in AI-generated content. For enhanced clarity, the watermark will be applied through a deep learning model.

"The invisible watermark is applied with a deep learning model. While it’s imperceptible to the human eye, the invisible watermark can be detected with a corresponding model. It’s resilient to common image manipulations like cropping, color change (brightness, contrast, etc.), screen shots and more. We aim to bring invisible watermarking to many of our products with AI-generated images in the future," the company added.

It should be noted that last month, Minister of State for Electronics and Information Technology, Rajeev Chandrasekhar, highlighted that the regulation of AI and ensuring user safety should not stall the growth of research, startups and the innovation ecosystem linked to this technology. Also, China's Cyberspace Administration has mandated regulations compelling generative AI vendors to label generated content, including text and image generators, while ensuring it doesn't disrupt user functionality. Also, Senator Kyrsten Sinema has highlighted the importance of transparency in generative AI during a recent US Senate committee hearing, suggesting the use of watermarks.