Meta commits to transparency with a new labeling approach for AI-Generated content

Meta this week announced major updates to its policy on AI-generated content, following extensive consultations and feedback from experts. The social media giant will now focus on increased transparency, labeling these materials rather than primarily removing them. This shift aims to balance freedom of expression and the fight against misinformation.

Key Changes:

• "Made with AI" Labels: Meta will introduce the "Made with AI" label for videos, images, and audio content that are either AI-generated or significantly AI-altered. This applies whether the AI signals are detected by Meta or the creator self-discloses the content as being AI-generated.
• Additional Context and High-Risk Labels: Content deemed to have a high risk of public deception on significant matters may receive a more prominent label, giving users greater context.
• Focus on Transparency, Not Removal: Acknowledging concerns over freedom of expression, Meta will prioritize labeling AI content. Removal will be reserved for cases where the content clearly violates existing community standards (e.g., hate speech, incitement of violence).

"A majority of stakeholders agreed that removal should be limited to only the highest risk scenarios where content can be tied to harm, since generative AI is becoming a mainstream tool for creative expression," wrote wrote Monika Bickert, Vice President of Content Policy at Meta. "This aligns with the principles behind our Community Standards – that people should be free to express themselves while also remaining safe on our services."

Why This Matters:

• Combating Misinformation: The new approach gives users tools to better assess the authenticity of the content they see.
• Balancing Free Speech: Prioritizing labels helps minimize potential restrictions on creativity and legitimate uses of AI tools.
• Evolving Policy: The updated policy reflects the ever-changing technological landscape.

Meta plans to implement these changes in May 2024. The company promises to continue working with industry partners, governments, and civil society to refine its approach towards AI-generated content.