Meta tests AI-generated images of users to boost interest in its AI tools

Meta is testing a new way to promote its generative AI tools by creating custom images of users based on their selfies. These AI-generated depictions are then shown to users within their Instagram feeds.

The platform’s AI uses uploaded selfies to generate imaginative versions of users in various scenarios. For example, one Reddit user, Green_Video_9831 shared that after editing their selfie with Meta AI, they saw their modified image being used in targeted ads.

Meta has reportedly confirmed this as part of a test to highlight its AI capabilities. However, this approach has sparked mixed reactions.

While Meta likely hopes to showcase the creativity of its AI tools, the results can feel strange or unsettling. The AI-generated images resemble users but raise questions about privacy and user consent. Some users may not appreciate seeing altered versions of themselves used in unexpected ways. One user replied to the original post noting that they had removed their photos from Facebook and reduced their permissions. 

While this experiment is concerning, it aligns with Meta’s broader AI strategy. The company envisions a future where users can interact with customised AI avatars and create multiple AI versions of themselves. These features are part of Meta’s push to integrate generative AI into its platforms.

CEO Mark Zuckerberg recently shared that the platform’s AI chatbot has nearly 600 million monthly active users. Meta has also invested heavily in AI infrastructure and expanding its data centres.

As the platform continues to push the boundaries of AI, it remains to be seen whether these innovations will enhance user experiences or make platforms feel less personal.