Meta has launched a new artificial intelligence model called Muse that enables users to generate AI content using public Instagram posts and reels. The feature is enabled by default, raising immediate privacy concerns. Users can even @-mention specific Instagram accounts within the Meta AI app to bring those profiles directly into generated images, according to a company announcement.
The privacy implications are significant: any public Instagram photo or reel can now be used as source material for AI-generated content without explicit consent from the original poster. While Meta frames this as a creative tool for tasks like designing custom event invitations, the default opt-in approach means users must actively adjust their settings to opt out. This mirrors past controversies where Meta's default privacy settings drew regulatory scrutiny.
Technically, Muse operates as a generative AI model trained on publicly available Instagram content. The integration with Meta AI allows seamless referencing of public profiles through @-mentions, effectively turning any public Instagram account into a potential asset for AI generation. No details have been provided on how Meta handles content removal requests or whether users can retroactively protect past posts.
No specific vulnerabilities, CVEs, or system compromises were reported, as this is primarily a privacy and consent issue rather than a cybersecurity exploit. However, security researchers may flag the default enablement as a vector for harassment or unauthorized use of personal imagery, potentially violating GDPR or other data protection frameworks in regions with strict consent laws.
Meta has not yet announced any mitigation beyond allowing users to adjust their settings manually. The broader context involves ongoing tensions between AI innovation and user privacy, with regulators in Europe and elsewhere increasingly scrutinizing default data-sharing practices.