"Meta allegedly earns roughly $7 billion in revenue every year from these types of potentially fraudulent ads, which can include ads for crypto scams, counterfeit medical products, and unlicensed casinos, another late 2024 document states. In fact, the documents show that Meta predicted in late 2024 that it would earn about 10% of its overall annual revenue, or $16 billion, from advertising fraudulent products."
"The internal documents showed that Meta's moderation process when it comes to dangerous advertisements has plenty of room for error. According to the reports, Meta's AI systems will only block an ad request if its AI determines there is more than a 95% chance of the ad being fraudulent. Meta allegedly charges higher pricing for running these "higher-risk" ads which show some signs of potentially being a scam, but don't reach the 95% threshold."
Meta displays 15 billion higher-risk advertisements daily that show signs of potential fraud and generates substantial revenue from advertising on Facebook and Instagram. The company still derives the vast majority of income from online advertising despite ventures into augmented-reality hardware. Meta earns roughly $7 billion annually from potentially fraudulent ads and projected about $16 billion, or 10% of yearly revenue, tied to fraudulent or indirectly related scams, including direct messages on Instagram or WhatsApp. Estimates in May 2025 placed the platforms as involved in a third of successful US scams. AI moderation blocks ads only above a 95% fraud probability, and higher-risk ads are charged at increased rates while high-value advertisers can receive many strikes before bans occur.
Read at PCMAG
Unable to calculate read time
Collection
[
|
...
]