Google Maps: an AI blocked 100 million fraudulent modifications in 2021

Google Maps removed 100 million abusive edits last year. To identify abusive changes, the American group uses artificial intelligence enhanced with machine learning.

Google Maps, the most popular mapping service in the world, allows users to make changes to maps. In particular, Google encourages users to add a missing place on a map or to move a place that would be badly placed. Of course, smart guys take the opportunity to add fake information on Google Maps.

That’s why Google reviews all changes made by users before applying them. Review of changes is a colossal and extremely time-consuming task for Google teams. “Every day we receive around 20 million contributions from people using Maps. These contributions include everything from hours of operation and updated phone numbers, to photos and reviews.“explains Google in a blog post.

Read also: Internet users use Tinder and Google Maps to fight fake news about the war in Ukraine

Google explains how an AI monitors changes to Maps

To identify abusive modifications without monopolizing its teams, the American company relies on machine learning. This is a state-of-the-art technology allowing algorithms to learn without having to be programmed beforehand. The AI ​​thus evolves over its experiences without the slightest action on the part of a programmer. Google uses machine learning on several other of its services, such as Google Search and Google Photos.

Thanks to a combination of human moderators and artificial intelligence algorithms, Google managed to prevent more than 100 million abusive changes on Maps. These fraudulent changes were intended in particular to tarnish the reputation of a business listed on the mapping application or to advertise it.

“We have identified and removed over 7 million fake business profiles […]We stopped more than 12 million attempts to create fake business profiles and nearly 8 million attempts to claim business profiles that didn’t belong to them”details Google.

The digital giant also specifies that it has blocked or deleted over 95 million photo edits violating its usage policy, more than 190 million photos and 5 million videos that are blurry or in violation of its rules. According to Google, the artificial intelligence deployed to protect Maps continues to evolve and is increasingly effective.