Google: Do not put AI labels in corners

Google has updated its help document on labeling AI content in advertising. The policy covers ads that were created or subsequently edited with artificial intelligence. Especially relevant is the new guidance on placing your own labels on image assets: labels should not be positioned in the very corners of the creative. With this change, Google refines a transparency rule introduced on July 9 that already affects advertisers, agencies, and SEO as well as marketing teams.

Background on AI labeling in Google Ads

When Google announced labeling for AI-assisted ads, the focus was primarily on clarity for users. Ad images and videos that were generated or heavily altered with generative models should be clearly recognizable as such. The help document “Use AI content labels” explains how advertisers should meet this requirement and which design standards apply. The latest addition now specifically addresses the placement of self-added labels on image materials.

For many campaigns, this is not a side issue. Image assets carry message, brand, and call to action. At the same time, platform elements, system notices, and logos often occupy exactly the edge and corner areas. If a custom AI label sits there, it can be covered, cropped, or hard to read. Google therefore explicitly warns against placing labels in the very tight corner zones.

Why image corners are problematic

In practice, interfaces and overlay components frequently claim the corners of creatives. That applies to ad previews, mobile feed layouts, and responsive formats as well as to landing-page components where assets are reused. A label in the corner then competes with UI elements and loses visibility. From a policy perspective, this is critical because transparency only works when the label can actually be noticed.

In addition, image crops vary by placement and device. A label that still looks visible on desktop may partly disappear on a smartphone. Corner areas are especially vulnerable during cropping. Placing labels outside these risk zones reduces the chance that the disclosure is missing or illegible in individual impressions.

Design and technical consequences

• Labels should sit in calm image zones with sufficient contrast.
• Corner areas should be kept free for system overlays and logos.
• Responsive crops and mobile formats must be considered during placement.
• Readability takes priority over purely aesthetic edge placement.

Relevance for SEO, content, and paid-search teams

Even though the rule originates in an advertising context, it also affects SEO and content workflows. Many teams reuse the same visuals in ads, social creatives, and on-page assets. When AI-generated or AI-edited images are used multiple times, labeling logic should stay consistent. Clear label placement supports compliance and helps prevent campaigns or asset approvals from being delayed because of unclear disclosure.

For technical SEO and analytics owners, the update is a signal to review workflows: Who creates the creatives? Which tools generate or edit images with AI? Where is it documented that an asset requires labeling? And who checks before upload that labels are not sitting in the corners? Such checkpoints can be integrated into existing approval processes.

From an E-E-A-T and brand-trust perspective, visible transparency also matters. Users should be able to recognize when content was machine-generated or heavily altered. A readable label can reduce confusion and strengthen the traceability of advertising messages.

Practical implementation in asset production

Teams should adapt their design templates. Instead of placing labels in a corner by default, a defined safe zone inside the creative is preferable—for example in the lower or side third, but with distance from the edge. Contrast, type size, and background area must be chosen so the label remains readable even under compression and in small placements.

It is also useful to separate platform-side labeling from custom labels. If Google itself places notices, custom additions must not overlay them. The updated help makes clear that advertisers take responsibility for placement as soon as they add their own markings.

Recommended review flow before upload

• Check whether the asset was created or edited with AI and requires labeling.
• Adapt label text and language to the audience and campaign requirements.
• Set a position outside the outermost corners and test against crop variants.
• Control contrast and readability on desktop and mobile.
• Document approval and tag the asset correctly in the DAM or campaign tool.

Context within Google’s current policy landscape

The change fits into a broader development: Google is regulating the use of generative AI in advertising more tightly and demanding more openness toward end users. After the July 9 announcement comes an operational fine-tuning. For advertisers, that means reading policy documents regularly and updating creative standards promptly.

Teams running international campaigns should check whether local teams apply the same placement rules. A central rule in the brand or ads playbook prevents inconsistencies. From an editorial and SEO perspective, the news mainly matters as a policy update. It does not affect the classic organic ranking algorithm, but it does touch the interface between paid media, asset quality, and transparency rules.

Overall, the update shows that Google follows up AI advertising rules with concrete design details. The message to advertisers is clear: custom AI content labels do not belong in the outermost corners of image assets. Teams that implement this now reduce compliance risk and ensure transparency notices reach users where they can actually see them.