Google updated its advertising policies on July 9, 2026, giving advertisers a direct way to disclose AI-generated or AI-edited creative inside campaign tools, rather than relying solely on invisible watermarking or after-the-fact review. The change, described in a changelog entry titled "Updates to AI labeling requirements (July 2026)," introduces what Google calls the AI label setting, a control rolling out gradually through July across five products: Google Ads, Display & Video 360, Campaign Manager 360, Merchant Center and Ads Editor.

The update responds to a regulatory reality that Google's own documentation states plainly. Rules in the European Union, India and New York require that ads containing certain AI-generated or AI-edited assets carry disclosures or labels informing consumers that AI played a role in their creation. Google's help pages frame the new setting as one of two compliance routes: advertisers can add labels directly to their creatives using their own design tools, or they can flip a setting inside the ad platform itself and let Google attach the label programmatically. Neither route, Google cautions, guarantees compliance with any specific regulation on its own; the documentation urges advertisers to seek legal guidance and confirm that their ads and assets meet local obligations.

What follows is a detailed look at how the mechanism works, and, product by product, the exact steps Google has published for turning the setting on.

How the AI label setting works

Two disclosure layers sit behind the feature, and they do different jobs. The first is global and largely invisible during normal browsing. Any ad containing an asset that an advertiser designates as AI-created or AI-edited will carry a disclosure inside the "How this ad was made" section of My Ad Center, the consumer-facing panel reachable through the three-dot menu on ads across Search, YouTube and Discover. Because this layer exists everywhere Google serves ads, a shopper in Toronto and a shopper in Mumbai can both, in principle, click through and see the same underlying disclosure, even though only one of them will see anything different about the ad's surface.

That second layer is regional. For campaigns targeting the European Union, India or New York specifically, assets an advertiser flags as AI-created or AI-edited also trigger a visible overlay directly on the ad, not just inside the click-through panel. The identical creative can therefore look different depending on where it runs.

A separate mechanism matters for anyone assuming labeling stays entirely optional. Google's documentation notes three circumstances under which it may apply a label without advertiser input: when Google itself is legally required to in a given region, when it receives signals from other platforms indicating AI involvement, or when an advertiser uses one of Google's fully automated tools to generate and serve creative. In each case, according to the documentation, "labels cannot be overwritten." An advertiser who never touches the AI label setting can still end up with a labeled ad, and there is no interface control to remove one applied this way.

Setting up AI labels across each product

Because the rollout spans five distinct interfaces, each with its own navigation logic, the mechanics differ noticeably from one to the next. Google Ads and Ads Editor share a similar structure built around an asset library and a "Review assets" prompt; Display & Video 360 leans on its Creatives gallery and bulk editing tools; Campaign Manager 360 and Merchant Center each fold the control into their existing creative-upload flows. Search Ads 360, documented separately, extends comparable controls to templates and individual search assets, even though it sits outside the five products named in the July rollout announcement.

Inside Google Ads, the "AI label" status lives as a column, visible from both the asset library and asset reporting, and advertisers can filter assets by that status and then apply sub-filters based on their own criteria. Three separate workflows exist depending on where in the campaign lifecycle an advertiser encounters an unlabeled asset.

During campaign creation, an in-product prompt to "Review assets" appears whenever ads in a campaign may require an AI label. From there, the sequence is short: select the relevant assets, choose Manage AI label, pick either "Label this asset as created or edited with AI" or "Don't label this asset," then select Done. For existing campaigns, the path runs through the Tools menu instead: Asset studio, then Asset library, then selecting an individual asset to reach its details page, applying the label choice and selecting Save. A third path appears specifically when new AI-generated assets are added to a campaign: a table lists assets with the "AI label" column still undesignated, and checking a box next to any of them surfaces the same labeling pop-up. Google's documentation adds two wrinkles worth noting here. If an advertiser selects an asset, such as an existing YouTube video, that already carries an AI label created outside Google Ads, the interface offers a link back to wherever that label originated so the advertiser can review it rather than resetting it. And uploading a video to YouTube directly from within Google Ads triggers a separate prompt asking whether AI was used, a question tied to YouTube's own organic-content disclosure policy rather than the ads-side setting.

Google Merchant Center

Merchant Center folds AI labeling into its existing image-management dialog. Inside a Merchant Center account, an advertiser opens "Manage images" within a product's details, selects the image or video assets that need attention, and finds "Manage AI label" tucked inside the three-dot menu at the top right corner of the asset. Choosing between the label and no-label options, then selecting Apply followed by Save or Add image, completes the sequence. The same control surfaces automatically when assets are generated through Product Studio or the platform's video assets pages, meaning merchants using Google's own generative tools for product photography encounter the labeling choice at the point of creation rather than as a separate cleanup task afterward.

Campaign Manager 360

Campaign Manager 360 attaches the AI labels dropdown to its standard creative-upload workflow rather than building a dedicated review screen. After opening a campaign or advertiser and selecting New to choose a creative type, an advertiser names the creative and sets its dimensions inside the "New creative" panel, then adds the file itself through the "Creative assets" section, either by selecting Add asset or dragging the file into the assets table. Saving that upload reveals an AI labels dropdown, where selecting "Label this creative as created or edited with AI" or "Don't label this asset," followed by a second Save, finalizes the choice.

Display & Video 360

Display & Video 360 offers the widest range of entry points, reflecting its role managing large creative libraries at scale. Advertisers can locate labeled and unlabeled assets through the Creatives gallery's filter bar, where an "AI label" filter breaks results into three states: on, off and not set. Bulk workflows come in two forms: a bulksheet with a dedicated "AI label" column for uploads, or, without a spreadsheet, selecting the field for a specific asset and choosing a label state directly, though creatives not hosted on Display & Video 360 will return an error rather than update. Editing already-published creatives in bulk runs through the "Edit multiple creatives" tool, applying a chosen label state across a selected group before a final Save that the documentation says can take roughly ten seconds per creative, meaning a large batch may take several minutes with the browser window kept open until it finishes. Fresh uploads carry the same choice under "Additional details," and assets pulled from the shared library, including unlabeled logos and images, go through a similar Review assets flow. YouTube video assets tied to ads get labeled through an SDF file or the API after upload, with SDF-based labeling available from version 10.1 onward.

Search Ads 360

Search Ads 360's documentation describes two related controls rather than a single unified flow, and it carries an important caveat about which search engines the labels actually reach. Templates, which govern how Search Ads 360 pushes changes into connected ad accounts, gain an "AI label" column reachable from a sub-manager account's Tools and settings menu, under Templates. Selecting a field in that column, choosing a label state and saving updates the template, though the change only takes effect once the template itself is enabled, a separate step involving checking the template's box, opening the Edit dropdown on the toolbar that appears, and selecting Enable; from that point, updates apply the next time the template runs. Individual assets can also be labeled directly from a client account's Assets page, where each asset detail view includes its own "AI label" section. The caveat concerns cross-engine reach: Search Ads 360's AI label fields function for Google Ads and Microsoft ad assets, but assets originating from other connected engines, such as Yahoo Japan or Baidu, default to a status of "Unknown" and remain there regardless of what an advertiser selects, since Search Ads 360 has no mechanism to push the label through to those platforms.

Ads Editor, the desktop application many larger accounts use for offline bulk management, mirrors much of the in-browser Google Ads workflow but adds bulk tools built around its shared asset library. Importing an existing AI image asset that lacks a label triggers a "Review assets" callout listing the relevant assets in a table, and checking the box next to one opens the same label-or-don't-label pop-up seen elsewhere; if the imported asset, such as a YouTube video, already carries a label created outside Ads Editor, the tool instead offers a link to review that existing label. Generating a new AI image asset inside Ads Editor follows a near-identical callout, but routes through a Manage AI label button before the labeling pop-up and a final Apply and Done. For assets already sitting in the shared library, selecting Shared library, then Asset library, then an asset type such as Images or Videos, surfaces the "AI label" column directly in the data view; opening an individual asset and selecting the pencil icon beside its AI label row brings up the same choice. Bulk labeling works the same way at scale: selecting Images inside the Asset library, checking every asset that needs attention, then using that same pencil icon applies the chosen label state across the whole selection in one pass.

Applying a custom label instead

Advertisers who prefer not to rely on the platform-generated disclosure can build their own label using external design tools and attach it to the creative directly. Google's documentation is explicit that a custom label of this kind will not run afoul of the company's separate policies banning text overlays and watermarks on ad creative, provided it functions as a disclosure rather than as decorative or misleading text.

Two practical constraints follow from how Google's systems process image assets after upload. First, ad rendering can trim the outer edge of an image, so a label placed too close to the border risks being cut off entirely; Google's guidance specifies that advertisers should "not place labels within a 5.5 percent margin around the full perimeter" of the asset to avoid this outcome. Second, Google's own image-enhancement features, which automatically crop and adjust assets to fit different placements, can crop a custom label out of frame without warning. Opting out of those enhancements avoids the risk, and the specific control differs by campaign type: Asset Enhancements governs the behavior in both Display and Performance Max campaigns, while Adaptive Layouts serves the equivalent function in Demand Gen campaigns.

Machine-readable metadata: SynthID and C2PA

Beneath the visible labeling layer, Google applies a second, non-visual disclosure mechanism to any asset generated through its own AI tools. That mechanism embeds SynthID, a non-visible digital watermark, alongside markup conforming to the C2PA standard, a specification maintained by the Coalition for Content Provenance and Authenticity that attaches cryptographically signed metadata describing an asset's creation and editing history. Both are woven into images and videos generated within Google Ads tools automatically, independent of whatever an advertiser selects in the visible AI label setting. Because this metadata persists inside the file rather than depending on advertiser self-reporting, it functions as a check that operates even when a person handling the creative later strips out or ignores the interface-level label.

Election advertising keeps a separate path

The changelog entry is careful to distinguish this new setting from an existing obligation that continues unchanged. Election advertisers must still disclose any Election Ad containing synthetic or digitally altered content by selecting the checkbox in the "Altered or synthetic content" section of their campaign settings, a requirement that sits inside Google's Political content policy rather than the newly launched AI label setting. Google also updated its Image assets format requirements policy article to reference the broader labeling change, without altering the separate election-content disclosure mechanism itself.

Why this matters

The rollout does not appear in isolation. Merchant Center required a narrower form of AI-image transparency before: in February 2024, Google mandated an IPTC metadata tag on AI-generated product images, an approach the new "Manage AI label" workflow now largely supersedes. SynthID itself dates to 2024 as well, when DeepMind introduced the watermarking system; Google later built a consumer-facing verification tool into Gemini, letting anyone upload a video and ask whether it carries the mark.

The creative tools generating the assets now subject to labeling have grown quickly, too. Google formally launched Asset Studio as a centralized creative-generation hub inside Google Ads, the same interface named in the platform's AI-labeling instructions. On the developer side, the Google Ads API added dedicated synthetic-content labeling structures on June 24, 2026, a programmatic equivalent published roughly two weeks before this changelog entry.

Nor is advertising the first Google property to face this question. YouTube introduced mandatory AI content disclosure for creators in May 2025, establishing much of the automatic-detection logic, including non-adjustable labels, that now reappears in the ads context.

None of this unfolds apart from advertiser sentiment. Research the IAB published in January 2026 found a wide gap between how advertisers and consumers feel about AI-generated ads, with far more executives assuming audiences view AI-made ads positively than consumers who actually report feeling that way. A labeling mechanism does not close that gap by itself, but it does shift disclosure from something advertisers might volunteer into something enforced at the interface level in three markets, sometimes automatically.

For teams managing creative across several products at once, the harder task may be consistency rather than mechanics. An asset labeled inside Google Ads does not inherit that status inside Display & Video 360 or Campaign Manager 360 if uploaded separately into each, since every product keeps its own "AI label" field. Reconciling that status across platforms, more than any single button, looks like the bulk of the real compliance work here.

Timeline

  • February 19, 2024: Google requires an IPTC metadata tag on AI-generated product images inside Merchant Center, an earlier and narrower transparency requirement.
  • May 21, 2025: YouTube's mandatory AI content disclosure policy for creators takes effect, establishing precedent for automatic, non-adjustable labeling elsewhere at Google.
  • December 18, 2025: Google adds SynthID verification to the Gemini app, letting users check whether video content carries the watermark.
  • June 24, 2026: The Google Ads API releases version 24.2, adding synthetic-content labeling structures at the programmatic level.
  • July 9, 2026: Google publishes "Updates to AI labeling requirements (July 2026)," introducing the AI label setting for Google Ads, Display & Video 360, Campaign Manager 360, Merchant Center and Ads Editor.
  • Throughout July 2026: The AI label setting rolls out gradually across the five named products.

Summary

Who: Google, through updates published across its Google Ads, Merchant Center, Campaign Manager 360, Display & Video 360, Search Ads 360 and Google Ads Editor help documentation, affecting advertisers and merchants running campaigns on any of those products.

What: An AI label setting that lets advertisers disclose AI-generated or AI-edited creative assets directly inside campaign tools, surfacing that disclosure globally through My Ad Center's "How this ad was made" panel and, for campaigns targeting the European Union, India or New York, through a visible overlay on the ad itself. Google may also apply labels automatically in certain cases, and those automatically applied labels cannot be removed.

When: Google posted the changelog entry describing the update on July 9, 2026, with the setting rolling out gradually across five products throughout July 2026.

Where: The setting applies globally for the My Ad Center disclosure layer, with the visible on-ad overlay restricted to campaigns targeting the European Union, India and New York specifically.

Why: Regulations in those three jurisdictions require disclosure when ads contain certain AI-generated or AI-edited assets, and Google's documentation frames the new setting as one available route toward meeting that requirement, alongside custom labels advertisers can build themselves, while explicitly declining to guarantee that using the setting satisfies any specific law on its own.