WhatsApp is developing a new generative AI feature that will allow users to create personalized avatars of themselves for use in various settings. The feature, spotted in the WhatsApp Beta for Android 2.24.14.7 by WABetaInfo, will use user-supplied images, text prompts, and Meta’s AI Llama model to generate the avatars. Users will be able to imagine themselves in settings ranging from the forest to outer space.
To create the avatar, users will need to take photos of themselves which will be used to train Meta AI. They can then generate their avatars by typing commands in the Meta AI chat or in other WhatsApp conversations. The feature will be optional and users will need to manually enable it in the settings. Reference images can be deleted at any time.
There is no specific timeline for the launch of this feature, as WhatsApp is still rolling out support for its Meta AI chatbot and real-time AI image generation in the US. Meta is likely taking a cautious approach given past issues with generative AI tools. The example images provided give an idea of what to expect from this new feature.
Source link
Source link: https://medium.com/@tankisompela010/whatsapp-to-introduce-ai-powered-avatar-creation-tool-e857d7ebdc99