In a significant move, Meta Platforms (META.O) has announced a forthcoming requirement for advertisers on Facebook and Instagram to disclose the use of artificial intelligence (AI) or other digital methods in creating or altering political, social, or election-related advertisements starting in 2024.
Meta Addressing Misrepresentation in Ads
According to a blog post from Meta, the disclosure mandate will specifically address scenarios where ads portray real individuals engaging in actions or making statements they did not, or where digitally rendered personas that do not exist are featured.
The company is also urging advertisers to reveal whether their ads depict events that did not occur, manipulate footage of actual events, or present a real event without utilizing the authentic image, video, or audio recording.
Prohibiting Political Use of Generative AI Ads
These policy updates follow Meta’s recent decision to prohibit political advertisers from employing generative AI ad tools. The platform had previously expanded access to AI-powered advertising tools, enabling advertisers to swiftly generate backgrounds, adjust images, and create variations of ad copy based on simple text prompts.
Notably, Alphabet’s Google (GOOGL.O, the leading digital advertising company, introduced similar image-customizing generative AI ads tools last week. Google aims to maintain a politics-free environment by blocking a set of “political keywords” from being used as prompts.
Growing Concerns from U.S.
Concerns from U.S. lawmakers about the use of AI to fabricate content that inaccurately depicts candidates in political advertisements have grown, given the accessibility and affordability of advanced “generative AI” tools capable of producing convincing deepfakes.
To address this, Meta has been proactive in blocking its user-facing Meta AI virtual assistant from generating photo-realistic images of public figures. Nick Clegg, Meta’s top policy executive, emphasized the need for rule updates in the realm of generative AI in political advertising.
It’s essential to note that Meta’s new disclosure policy won’t apply to digital content deemed “inconsequential or immaterial to the claim, assertion, or issue raised in the ad.” Examples include adjustments such as image resizing, cropping, color correction, or image sharpening.
This initiative underscores Meta’s commitment to enhancing transparency and curbing the potential misuse of AI in the realm of political advertising, aligning with broader industry efforts to navigate the challenges posed by rapidly advancing technology.
Read More (AI)
