A new investigation alleges that Meta prioritized billions in revenue from scam advertisements over user safety, internally calculating that potential fines were cheaper than cracking down.
In a revelation that casts a long shadow over the social media giant's public safety commitments, a trove of internal Meta documents suggests the company failed to effectively curb a deluge of fraudulent advertisements on Facebook and Instagram for at least three years. The documents, reportedly dating from 2021 to 2025, form the basis of a major investigation that paints a picture of a company aware of the scale of the problem but unwilling to sacrifice the substantial revenue it generated.
According to the report, Meta's automated systems display approximately 15 billion ad impressions per day that show clear signs of being high-risk or fraudulent. These ads, which often promote cryptocurrency scams, fake merchandise, and other deceptive schemes, are alleged to generate a staggering $7 billion in annual revenue for the tech behemoth.
A Calculated Business Risk: Penalties vs. Profit
The internal documents reportedly reveal a core reason for this failure: a deliberate and hesitant approach to enforcement. Sources indicate that if Meta's automated systems detected a fraudulent ad with less than 95% certainty, the company would not ban the advertiser. Instead, it would levy a "penalty bid"—effectively charging the scammer a higher rate for their ad, a cost often easily absorbed given the high returns from successful fraud.
Perhaps the most damning allegation is that Meta engaged in a ruthless internal cost-benefit analysis. The Reuters investigation found that company officials estimated potential regulatory fines for failing to police these ads would be a maximum of $1 billion—a figure significantly less than the estimated $3.5 billion in annual profit from the high-risk ad segment. This has led to a strong suspicion that Meta consciously operated on the principle that paying occasional fines was a cheaper option than forfeiting a multi-billion dollar revenue stream.
AI Ambitions Funded by Fraud, Security Teams Handcuffed
The report further suggests that the massive revenue from these fraudulent ads is being used to bankroll CEO Mark Zuckerberg's expensive ambitions in the artificial intelligence (AI) arms race, with a planned capital investment of up to $72 billion.
Meanwhile, the company's security and fraud prevention teams were allegedly operating with severe financial handcuffs. In the first half of 2025, the team responsible for stopping scams was reportedly forbidden from implementing any measures that would cost Meta more than 0.15% of its total revenue—approximately $135 million—a drop in the bucket compared to the revenue generated by the very ads they were trying to stop.
Users Left to Pay the Price
The consequences of this alleged prioritization are said to fall directly on users. In 2023, Meta is accused of ignoring or incorrectly rejecting a shocking 96% of the roughly 100,000 valid fraud reports submitted by users each week.
The human cost is starkly illustrated by the case of a hacked U.S. Air Force recruiter's Facebook account. Through this compromised account, a colleague was allegedly scammed out of $28,000 in a cryptocurrency fraud—a case that underscores the real-world harm caused by what internal documents reportedly describe as negligent security standards.
Meta Fights Back, Disputing the Allegations
In response to these allegations, Meta spokesperson Andy Stone issued a firm denial. Stone stated that the documents presented a "selective view" that distorted the company's complex work in this area. He contested the internal estimate that 10.1% of revenue came from prohibited advertising, calling it a "rough" figure that was too high because it included "many" legitimate ads.
Stone emphasized that Meta "aggressively combats fraud," pointing to the company's own metric that it has reduced user reports of fraudulent ads worldwide by 58% as evidence of its commitment and effectiveness.