Google AI blocked 3X more advertising fraud in 2024

Google‘s increased use of AI models to combat fraudulent advertising has achieved unprecedented results, with suspended accounts tripling and deepfake scam ads plummeting by 90% in 2024. This application of large language models (LLMs) represents one of the most broadly beneficial implementations of AI technology to date, showing how advanced models can be deployed to protect users from digital threats while maintaining advertising ecosystems.

The big picture: Google deployed over 50 enhanced LLMs to enforce its advertising policies in 2024, with AI now handling 97% of ad enforcement actions.

• These models can make determinations with less data than previous systems, enabling faster responses to rapidly evolving scam tactics.
• The company claims most fraudulent accounts were caught before running a single ad, representing a significant improvement in preventative measures.

By the numbers: Google suspended 39.2 million US ad accounts for fraudulent activities in 2024, more than triple the 12.7 million accounts suspended in 2023.

• The company removed 5.1 billion bad ads globally (1.8 billion in the US), slightly down from 5.5 billion in 2023.
• This reduction in removed ads despite increased enforcement suggests Google’s preventative approach is working, stopping fraudulent accounts before they can distribute malicious content.

Combating AI misuse: A specialized team of 100 experts helped update Google’s misrepresentation policy to address emerging AI-enabled threats.

• The updated policies led to blocking 700,000 advertiser accounts, resulting in a 90% reduction in deepfake scams in advertisements.
• Google also blocked 1.3 billion pages from showing ads in 2024, primarily due to sexual content, dangerous or derogatory material, and malware.

Why this matters: The effectiveness of Google’s AI-powered enforcement shows how LLMs can be applied to solve real-world problems that benefit broad user populations, potentially appealing even to AI skeptics.

The human element: Despite the extensive automation, human oversight remains part of Google’s process to address potential false positives and account suspension errors.

• The dramatic increase in account suspensions creates greater opportunity for legitimate accounts to be incorrectly flagged, though Google maintains the overall impact has been positive.