Google suspended 39M advertiser accounts last year, thanks to AI

Throughout 2024, Google suspended over 39.2 million advertiser accounts on its platform, which is more than three times the number compared to the year before.

Cybercriminals and other bad actors constantly adapt their tactics to evade detection, making it harder and harder for tech companies like Google to expose ad fraud. In recent years, Google has introduced artificial intelligence (AI) as a tool to detect ad scams and distinguish legitimate businesses from fraud.

Google’s latest generation of large language models (LLMs) operates much more efficiently. Compared to earlier models, they only need a fraction of the information to recognize and identify fraudulent ads or ad campaigns.

Last year, Google introduced over 50 enhancements to its models to detect ad fraud. This move has paid off.

According to Google’s 2024 Ads Safety Report, the malicious use of its ad platforms has significantly decreased. The report states that Google suspended over 39.2 million advertiser accounts, removed over 5.1 billion ads, and restricted another 9.1 billion ads.

Another tool to combat bad actors is Google’s Advertiser Identity Verification. This process is designed to confirm the real identity of advertisers before they can publish ads on Google’s platforms in order to prevent fraud and increase transparency. The program covers more than 200 countries and territories and makes sure that, on average, 90% of all ads seen by people on Google come from verified advertisers.

Bad actors using AI-generated images to imply an affiliation with a celebrity to promote a scam is an industry-wide problem. To fight back, Google founded a team of over 100 experts to analyze these scams and develop effective countermeasures. As a result, Google suspended over 700,000 offending advertiser accounts.

“These preventative efforts kept billions of policy-violating ads from ever showing to a consumer while simultaneously ensuring that legitimate businesses can show ads to potential customers quickly,” the Ads Safety Report reads.

It continues by saying that these technical advancements allow Google’s teams “to focus on more complex, ambiguous problems, which in turn provides our LLMs with nuanced training data to better address these instances in the future.”

Google promises to continue working on ad scam prevention in the future.