Meta Plans To Automatically Label AI-Generated Content

Meta

In the coming months, Meta plans to automatically label AI-generated content uploaded to Facebook, Instagram, and Threads. The company is collaborating with industry partners to develop a common technical standard for identifying AI content, enabling Meta to detect and label AI-generated content from various sources like Google, OpenAI, Adobe, and Midjourney.

Initially, the focus will be on photos, as AI tools for videos and audio lack sufficient signals for accurate detection. For the time being, users will be required to disclose and label this type of content, potentially facing penalties if they fail to do so.

Why It Matters: Meta's efforts are relevant in light of upcoming elections, where AI-driven misinformation is expected to be an issue. By taking steps to proactively label digitally altered or manipulated content, Meta can provide more transparency and protection for users.

TikTok announced a similar move last year, but its effectiveness remains unknown, as minimal content is labeled as AI despite issues with deepfakes. Meta is likely to fare better, but identifying and labeling AI content will require consistent efforts. Creators will need to play a role as well, ensuring they properly disclose AI content, similar to sponsored content.