Google announced it has new AI and large language models to fight invalid ad traffic or as Google calls it "IVT." Google explained they have used AI for a long time, but they had some breakthroughs that resulted in a "40% reduction in IVT stemming from deceptive or disruptive ad serving practices."
Google wasn't too specific about what is new here but they did say that the Google Ad Traffic Quality team, with Google Research and Google DeepMind, "recently introduced industry-leading defenses powered by large language models to more precisely identify ad placements generating invalid behaviors."
These "new applications provide faster and stronger protections by analyzing app and web content, ad placements and user interactions," Google added.
Invalid traffic is any activity that doesn't come from a real user with genuine interest. It can include accidental clicks caused by intrusive ad implementations, fraudulent clicking by competing advertisers, advertising botnets and more.
Ginny Marvin, the Google Ads Liaison, added on X:
We’re using AI in new ways to fight invalid traffic (IVT). Here’s how:Working with Google Research and Google DeepMind, we've developed new LLM-powered solutions that allow us to more precisely identify ad placements that are generating interactions from non-genuine users.
This significantly improves our ability to prevent ad budget waste on IVT and exclude policy violators from our platforms.
These new solutions have already reduced IVT stemming from harmful ad serving practices by 40%.
Learn more about our Ad Traffic Quality team’s work to combat invalid traffic and ensure ad spend goes to reaching real people with genuine interest.
Forum discussion at X.
