Abstract：AI takes the lead in ad screening, blocking billions of harmful ads and safeguarding online safety worldwide.
Amid rising concerns over online fraud and misinformation, Google's 2024 Ads Safety Report reveals that artificial intelligence has become a frontline defense against malicious advertising. Powered by advanced models like Gemini, Google blocked over 5.1 billion harmful ads, restricted the reach of 9.1 billion, and suspended more than 39 million advertiser accounts, setting new records in enforcement efficiency.
Online scams are becoming more sophisticated, from AI-generated content to impersonation of public figures. Google's AI now detects signs of fraud—such as stolen payment info, fake businesses, or coordinated scam networks—more swiftly and accurately than ever.
In regions like Africa, misleading political ads and impersonation scams remain serious issues. To combat this, Google strengthened its misrepresentation policy and deployed a global review team of over 100 experts, resulting in the shutdown of more than 700,000 scam-related advertiser accounts. This led to a 90% drop in impersonation scam reports.
With nearly half the world heading to the polls in 2024, Google also removed over 10 million election-related ads that failed to meet transparency requirements for identity verification and sponsorship disclosure.
Evolving Threats & AI‘s Limits
Despite AI’s impressive capabilities, the fight against evolving online threats continues. As scammers grow more inventive—especially in areas like political manipulation and deepfake content—AI systems face increasing pressure to adapt.
Alex Rodriguez, Googles General Manager of Ads Safety, noted:“We rolled out more than 50 AI model upgrades in 2024 to act faster and smarter—stopping threats before they reach users.”
However, he emphasized that human reviewers still play a key role in handling complex cases that AI cannot fully resolve.
Looking ahead, the collaboration between AI, human oversight, regulators, and global anti-fraud alliances will be essential. Technology alone is not enough—responsible frameworks and real-world judgment are equally vital in safeguarding the digital space.
