WhatsApp is reportedly working on a new generative AI feature allowing users to create personalized avatars for use in various imagined settings. This upcoming feature, known as "Imagine," was discovered in the latest WhatsApp Beta for Android 2.24.14.7 by WABetaInfo. It utilizes a combination of user-supplied images, text prompts, and Meta’s AI Llama model to generate these avatars.
According to a screenshot of WABetaInfo, users can envision themselves in any setting, from a forest to outer space. This functionality mirrors the capabilities of other AI generators like Lensa AI and Snapchat’s “Dreams” selfie feature.
To generate these personalized avatars, users must take photos of themselves, which will train Meta AI to replicate their likeness in different settings. Users can create avatars by typing “Imagine me” and describing the desired settings in Meta AI chat. Alternatively, they can use the feature in WhatsApp conversations by typing “@Meta AI imagine me...”
The feature is expected to be optional, requiring users to enable it manually in the WhatsApp settings. Users will also be able to delete their reference images at any time via the Meta AI settings.
The exact release date for the Imagine AI feature is currently unknown. WhatsApp is still rolling out its Meta AI chatbot and real-time AI image generation for users in the US. Given the complexities and previous issues with generative AI tools, Meta is cautiously deploying this new feature.