Meta has updated the disclosure labels that appear on Facebook and Instagram ads containing artificial intelligence generated content, aligning these AI specific notices with the broader changes the company recently made to its general ad disclosure system. The move is aimed at allowing users to learn which ads feature AI-generated content regardless of whether this content was created with Meta’s own services or third party artificial intelligence tools completely unrelated to Meta’s ecosystem.
The update matters because AI generated advertising content has become increasingly common across Facebook and Instagram, and until now, the way Meta flagged that content wasn’t always consistent with how it labeled ads more broadly. Meta is trying to unify its labeling practices to make its policies easier for the average user to understand. The company appears to recognize that AI generated visuals are becoming indistinguishable from unedited photos, making this kind of disclosure important for maintaining user trust in their feed.
Meta says that ads generated or modified with specific generative AI capabilities will receive the appropriate disclosure label in the About this ad field, available by clicking the three dot menu on each promoted post. In fact, it is one of the most reliable sources for users willing to learn about the origin of an ad, and the addition of an AI disclosure here makes sure that users will find this information where it is supposed to be.
How the New Labeling System Works
According to Meta, it will apply the AI info label if ads are created or edited using certain generative AI features like Background Generation, Image Generation, and Add Animation whenever these features are applied to significantly modify the images or videos. In other words, minor edits won’t require an AI label, and the key here is the extent to which the AI features affect the ad content.
This is valuable information for advertisers, since not every AI involved modification to an ad requires disclosure. The AI disclosure threshold is related to the degree to which these features influence the essence of the advertisement, and not the mere presence of artificial intelligence in the editing process. For instance, an advertiser applying minor AI-powered color corrections may have different results compared to another advertiser who uses AI to create a new background or add animation effects to his ad.
It should be noted that Meta doesn’t limit its labeling initiative only to its own AI tools since Meta says that it is going to label ads as created or edited with the help of the AI if it detects the presence of the relevant third-party tools like Photoshop’s generative features or Dall-E.
Detecting Third Party AI Content
In order to detect artificial intelligence in the ads created or modified with the help of the third-party services, Meta uses the established methods of detection, especially C2PA standard that is an abbreviation for the Coalition for Content Provenance and Authenticity.
C2PA inserts metadata into digital files disclosing how images and videos were created and edited, including the involvement of AI tools at any stage of this process. When the system detects this metadata, it applies the proper label automatically. It is essential since the advertisers cannot bypass the disclosure requirements by using AI tools different from the ones provided by Meta.
It should be noted that it is an ongoing process since Meta claims that it is going to continue improving the approach to the labeling of AI generated content while working alongside experts in the area, advertisers, policy stakeholders, and other industry participants considering the constantly changing public expectations regarding AI transparency and technological innovation.
Part of a Broader Shift in Ad Transparency
The update is not an isolated event, since it came just a few months after Meta announced an upgrade to its in-stream ad labels. In fact, Meta decided to abandon the word Sponsored in favor of Ad label. It was done in accordance with Meta’s efforts to simplify and standardize the labeling of the promotional content, and the recent labeling of AI ads looks like a continuation of this trend.
Actually, AI disclosure tags are already available for the organic, non-paid content. This means that users probably have seen similar AI labels on regular posts before this update was introduced for paid ads.
Why This Update Matters for Advertisers and Users
This development is useful to users browsing Facebook and Instagram feeds because it provides a straightforward and reliable way to determine if an ad used AI generated visuals. Given how sophisticated AI generated visuals are getting in their resemblance to photographs or videos, it is important to know if any ad used this technology so that one can consider the implications in the evaluation of the promotional material.
The importance of this update becomes evident in situations where accurate visuals are important, such as ads of products, food, travel destinations, etc. In these cases, there is a chance that some AI generated visual might mislead a consumer to purchase something they would not have if they knew the difference between AI generated images and authentic photos or videos.
For advertisers, this means that transparency is now an expected part of the process when generating ads, instead of an extra thing they can provide. Companies and brands that use a lot of AI generated visuals for their marketing should expect that the public will see information about whether the ads used AI or not. They may want to prepare their strategy for dealing with these labels accordingly.
Agencies working on paid advertisements on behalf of their clients should pay attention to the difference Meta makes between small AI assisted edits and larger use of AI tools. Since the label seems to depend on how much a particular AI tool changes the image or video, it will be wise to keep track of which AI features were used in the advertisement.
With generative AI being used in more and more advertisements, this labeling system is likely to keep evolving. The fact that Meta can detect if any ad uses generative tools other than the ones available on its platforms shows that the company is ready to create a labeling system that will keep up with how advertisers use AI technologies. It will be interesting to see whether this level of detection meets the standards set by the regulators and watchdogs calling for more transparency regarding AI usage.
Frequently Asked Questions
Q 1: What triggered Meta’s update to AI ad disclosure labels?
Meta updated the labels to align its AI specific ad disclosures with recent changes it made to its general ad disclosure system, creating a more consistent labeling experience across Facebook and Instagram.
Q 2: Where can users find the new AI disclosure label?
The label appears inside the About this ad section, which users can access by tapping the three dot menu on any promoted post on Facebook or Instagram.
Q 3: Which Meta AI tools trigger an automatic label?
Meta said it will automatically apply an AI info label when advertisers use tools like Background Generation, Image Generation, or Add Animation to significantly create or edit an image or video.
Q 4: Does Meta label ads made with third party AI tools?
Yes. Meta said it will also label ads created or edited using external generative AI tools, such as Photoshop or Dall-E, when it detects relevant metadata.
Q 5: How does Meta detect AI content from outside tools?
Meta uses industry standard detection methods, including the C2PA standard, which embeds metadata showing how an image or video was created or edited.
Q 6: Will every AI assisted edit result in a label?
No, not always. The label is based on whether the tool modifies the picture or video in a major way or only performs minor adjustments.
Q 7: Is this Meta’s first AI disclosure system?
No. Meta already had AI disclosure tags in place for organic, non-paid content before extending the same approach to paid ads.
Q 8: How does this relate to Meta’s Sponsored to Ad label change?
Meta has already updated its labeling policy once by switching from ‘Sponsored’ to ‘Ad’. This one is connected to that.
Q 9: Why does this update matter for users?
Users now have additional information that an ad contains some AI-generated visuals, which can be especially helpful when it comes to visuals of products, food, or travel.
Q 10: Will Meta’s approach to AI labeling keep changing?
Yes. Meta said it plans to continue refining its labeling approach in partnership with experts, advertisers, and industry partners as AI technology and public expectations evolve.
