Meta tests AI scanning of users' private phone photos for content suggestions
Facebook is testing an AI feature that analyzes photos from users' phone galleries to generate content suggestions, reportedly without clear user awareness of the scope of data being accessed. The system scans facial features, locations, and timestamps, with Meta's terms granting broad permissions for photo use and retention.
What happened
Facebook is testing a feature that analyzes photographs from users' phone galleries using artificial intelligence to generate suggestions for posts and stories, according to a cybersecurity expert warning on 8 July 2025. The feature, which appears as a prompt during post or story creation, requests users to enable 'cloud processing' to receive 'creative ideas created for you from your gallery.'
The system uses data including photo timestamps, locations, and content to generate compilations, summaries, collages, or AI-styled variations of images. Meta's terms of service for the AI service grant Meta broad permission to access, store, and analyze the shared photographs, including facial features, and to retain and use all shared information.
Reported impact
A cybersecurity expert noted that users often grant Facebook permission to use their photographs without awareness of the full scope, with the feature scanning gallery photos that have not been published, along with facial data, locations, and timestamps that users may not realize they have consented to share.
Index analysis
The feature was introduced with minimal notice, raising concerns among users about privacy implications. Cybersecurity experts advised users to verify privacy settings and restrict photo gallery access.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.