Google Removes 247 Million Ads, Suspends 2.9 Million Accounts In India For Policy Violations

Google removed over 247.4 million ads and suspended 2.9 million advertiser accounts in India for policy violations in 2024.

This information was revealed to the public in the 2024 edition of the data giant's annual 'Ads Safety Report.' The top five offenders of the company's advertisement policy in India were financial services, trademark violators, ad network abusers, personalised ads and gambling and games.

On a global scale, the software firm removed 5.1 billion ads that they deemed "harmful to users." The most common reason for ad removal was for "abusing the ad network," which was the reason 793.1 million ads were removed. This was followed by trademark violation, which got 503.1 million ads pulled.

The least common reason was listed as "counterfeit goods," which only led to the removal of 8,90,000 ads.

The firm also suspended 39.2 million accounts for "advertiser fraud".

Similarly, Google restricted 9.1 billion ads for "cultural and legally sensitive" reasons, reducing the reach of these ads to users. The firm restricted the most ads due to legal requirements (428.8 million) and for offering financial services (268.3 million).

The company reported using artificial intelligence-powered 'large language models' which flagged up upto 97% of the pages Google took action against for policy violation.

"By using these models, we significantly expedited site reviews, enabling quicker monetisation while keeping ads from appearing on violative pages," the report said.

The company also invested heavily in their LLM models in the previous year, implementing over 50 enhancements for precision and efficiency. This was to have them tackle problems that had more complexity and ambiguity. This provided it with more nuanced data to train on.

The search engine firm also said it was expanding its identity verification and transparency requirements for election advertisers to new countries.

These measures consisted of "paid for by" disclosures and a public transparency report of all election ads. It also claimed to be the first company to have disclosure requirements for AI-generated content in election advertisements.

The company reported that it took action against 1.3 billion web pages under its publisher policy requirements, with the leading reason being sexual content (1.2 billion), followed by dangerous and derogatory content (24.8 million).