This is how Google Maps is cracking down on spamming with AI


Google Maps is one of the most popular navigation apps used by many for finding directions, public transportation, and business information. Apart from official listings, people can also add videos, reviews, and ratings to the app, but sometimes this information can be incorrect or fake.
In the previous year, Google Maps introduced a major update that enables its machine learning models to detect new types of misuse much more quickly than in the past.
“For example, our automated systems detected a sudden uptick in Business Profiles with websites that ended in .design or .top — something that would be difficult to spot manually across millions of profiles. Our team of analysts quickly confirmed that these websites were fake – and we were able to remove them and disable the associated accounts quickly,” explains Google in a blog post.
The new machine learning model has been developed to prevent scammers from uploading images with fake phone numbers overlaid on them. These scammers aim to deceive unsuspecting victims into contacting them instead of the legitimate business. The machine learning model can detect such images by analysing their specific visual details and layout.
“Scammers started overlaying inaccurate phone numbers on top of contributed photos, hoping to trick unsuspecting victims into calling the fraudster instead of the actual business,” said Ashish Gupta, Google’s engineering director for user-generated content.
Gupta said in a blog post that the company used machine learning to eliminate millions of fake user-submitted content from Google Maps in 2022. This included 115 million policy-breaking reviews, 200 million low-quality photos, and 7 million inappropriate videos.
Google Maps also increased its efforts to prevent fake Business Profiles, with 20 million attempts blocked in 2022, an increase of 8 million from the previous year. To further combat evolving tactics of bad actors, Google also implemented safeguards for 185,000 businesses after detecting suspicious activity and attempts at abuse.


Source link

Leave a Comment