Meta to Require Disclosure of AI Use in Political and Social Ads
Date: 2023-11-08 Author: Dima Zakharov Categories: TECHNOLOGY
Meta Platforms, the parent company of social media giants Facebook and Instagram, is introducing a new advertising policy that will increase transparency and accountability in the digital advertising landscape. Starting in 2024, advertisers will be obligated to disclose their use of artificial intelligence (AI) or other digital techniques when altering or crafting political or social advertisements on the platform.
The policy will mandate advertisers to reveal whether their modified or generated ads feature real individuals who are shown doing or saying things they didn't actually do, or if they create non-existent digital personas that appear as real. Advertisers will also be required to signal if their ads depict events that never happened, involve manipulated footage of real events, or portray real events without authentic imagery, video, or audio recordings of those events.
Notably, the new policy will not require disclosure of information when digital content is "insignificant to the claim, statement, or issue being addressed in the advertisement." This includes regulations related to image size adjustments, image cropping, color corrections, or image sharpness enhancements.
This development follows Meta's previous declaration to prohibit political advertisers from using AI-powered advertising tools, and it comes just a month after the company expanded advertisers' access to AI-driven advertising tools capable of instantly creating backgrounds, modifying images, and varying ad text.
Meanwhile, Google recently unveiled similar AI-driven ad generation tools and stated its intention to keep its policy separate from its products, blocking a list of "political keywords" that cannot be used as prompts.
Additionally, Meta is offering European users the option to access Facebook and Instagram without ads for a subscription fee, as part of its ongoing efforts to enhance the user experience and provide an ad-free alternative.
The policy will mandate advertisers to reveal whether their modified or generated ads feature real individuals who are shown doing or saying things they didn't actually do, or if they create non-existent digital personas that appear as real. Advertisers will also be required to signal if their ads depict events that never happened, involve manipulated footage of real events, or portray real events without authentic imagery, video, or audio recordings of those events.
Notably, the new policy will not require disclosure of information when digital content is "insignificant to the claim, statement, or issue being addressed in the advertisement." This includes regulations related to image size adjustments, image cropping, color corrections, or image sharpness enhancements.
This development follows Meta's previous declaration to prohibit political advertisers from using AI-powered advertising tools, and it comes just a month after the company expanded advertisers' access to AI-driven advertising tools capable of instantly creating backgrounds, modifying images, and varying ad text.
Meanwhile, Google recently unveiled similar AI-driven ad generation tools and stated its intention to keep its policy separate from its products, blocking a list of "political keywords" that cannot be used as prompts.
Additionally, Meta is offering European users the option to access Facebook and Instagram without ads for a subscription fee, as part of its ongoing efforts to enhance the user experience and provide an ad-free alternative.
