WhatsApp is bringing AI-image creation to your phones and here's how it will work

Meta AI is introducing a new feature to WhatsApp that allows users to instantly generate images and videos directly within their chats. This new tool is part of a broader rollout of Meta AI features across the platform, now accessible in more than a dozen countries globally. However, the feature is still in the testing phase in India where only select users are getting the new feature. 

Real-time image and video generation

WhatsApp users can bring their imaginations to life in real time by simply typing into the chat. The AI instantly begins to generate images and videos that change with each word you type. This feature uses the latest Llama 3 large language model introduced by Meta.

How it works

The process is straightforward: as users type into their WhatsApp chat, whether describing a scene, an object, or any abstract concept, the AI starts to visualize it. The more details provided, the more refined the image becomes. This allows for a high degree of customization and creativity, making each generated image unique to the user's input. This is unlike the current image generation models like OpenAI's Dall-E 3 which on takes the final prompt from the user and then creates an image according to the instructions. 

User engagement and feedback

Meta claims that it is keen on collecting feedback from users to further refine and improve this tool. They encourage everyone to experiment with this feature and share their creations and experiences.

Meta Llama 3 roll out

Meta has also announced the release of its latest large language model, Llama 3 which supports the real-time image generator that will be offered on WhatsApp. Meta claims that this new upgrade will make it's AI offering the best among its free-to-use counterparts.