Google Blocked a Record 8.3 Billion Ads as AI Changed How It Enforces Policies

Google's ad safety numbers show a strategic shift

Google says it blocked 8.3 billion ads worldwide in 2025, a sharp increase from 5.1 billion the year before. At the same time, the company suspended fewer advertiser accounts than such a rise might suggest, a pattern that points to a broader change in how it is policing its ad systems.

According to the supplied report, Google attributes the shift to heavier use of AI, particularly its Gemini family of models. The company says those systems helped it detect and block policy-violating ads earlier and with greater precision, catching more than 99 percent of such ads before users saw them.

From bad actors to bad ads

The headline number is large, but the more revealing detail is the mismatch between blocked ads and suspended accounts. Rather than focusing primarily on removing entire advertisers, Google appears to be leaning more heavily toward stopping individual problematic ads on a case-by-case basis.

That approach suggests a more granular enforcement model. If AI systems can identify suspicious content patterns quickly enough, the platform can intervene earlier in the ad pipeline, sometimes before an advertiser account is suspended or before a campaign has significant reach.

Generative AI is raising the pressure

Google also says the increase in blocked ads reflects the growing use of generative AI by scammers. That matters because generative systems can produce deceptive text and creative variations at scale, allowing bad campaigns to mutate rapidly. In that environment, moderation tools that rely on manual review or slower rule updates can struggle to keep pace.

Google's answer is to use its own AI systems to detect patterns across campaigns and block them sooner. The report frames this as part of a larger push to integrate Gemini more deeply into Google's products and infrastructure, including advertising, where the company is using AI to automate campaign creation, enforce policies and respond to emerging threats in real time.

Scams remain a major category

The scale of scam-related enforcement remains significant. Google said 602 million blocked ads and 4 million suspended advertiser accounts were tied to scams. In the United States alone, the company removed more than 1.7 billion ads and suspended 3.3 million advertiser accounts in 2025. The most common violations there included ad network abuse, misrepresentation and sexual content.

India, Google's largest market by users, showed a similar but distinct pattern. The company blocked 483.7 million ads there, almost double the previous year, while account suspensions fell to 1.7 million from 2.9 million. The leading violations in that market included trademarks, financial services and copyright issues.

What the numbers suggest

The emerging picture is a platform shifting from blunt-force enforcement toward predictive screening at scale. That does not necessarily mean Google is going easier on advertisers. In some ways, blocking more ads earlier could be a stricter system because it interrupts campaigns before they gain momentum. But it does mean that account suspension is becoming a less complete proxy for overall enforcement intensity.

It also highlights how central AI has become to the health of the digital advertising ecosystem. The same generative tools that help marketers create campaigns faster are also making abuse easier to produce and harder to classify. Google's data suggests the company now sees AI not as an optional enhancement to moderation, but as the operating layer required to keep pace.

Whether that balance proves durable will depend on how well the systems distinguish between harmful and legitimate ads. But Google's latest safety report already makes one point clear: the enforcement battleground is moving down to the ad level, and AI is increasingly the mechanism that decides what gets through.

• Google says it blocked 8.3 billion ads globally in 2025, up from 5.1 billion in 2024.
• The company says more than 99 percent of violating ads were caught before users saw them.
• Google suspended fewer advertiser accounts despite the surge in blocked ads.
• The company links the change to deeper use of Gemini models in enforcement.

This article is based on reporting by TechCrunch. Read the original article.