In 2024, Meta announced a new approach to labeling AI-generated content. To enhance transparency, 'AI info' labels were added to images and videos generated or edited by AI.
This aims to help users better understand the context of content, but initial attempts led to confusion, prompting further adjustments.
Redesigning Labels to Meet User Expectations
Meta applies the 'AI info' label based on industry-standard AI image indicators and user self-reporting. However, the initial 'Made with AI' label faced criticism.
The label was applied even for minor AI edits, causing dissatisfaction among professional photographers using tools like Adobe Photoshop. To avoid misunderstandings, Meta has modified the label to be visible only in the post menu for simple AI edits.
Balancing Content Transparency and Freedom of Expression
Meta's new policy aims to ensure transparency while protecting freedom of expression. Moving forward, the 'AI info' label will be prominently placed only in cases with significant potential impact on people.
Additionally, content with a high potential for misunderstanding will undergo independent fact-checker reviews, with warning labels added as necessary.
Challenges Towards the Future of Labeling
Meta continues to collaborate with the industry to improve AI content labeling. The new labels will evolve with technological advancements, aiming to help users better understand AI-generated content.
This could enable many creators to safely expand their expressive range while leveraging AI technology.
Photo by hiro
