Recently, several internal Meta documents were leaked, revealing that approximately 10% of the social media giant's total annual revenue in 2024 (US$16 billion, or about RMB 114 billion) is suspected to have come from fraudulent activities and advertisements for prohibited products, sparking widespread concern.
In response, Meta spokesperson Andy Stone stated that the relevant documents were "one-sided" and that the 10.1% revenue share was a "rough and overly inclusive estimate."
"Clicking on fraudulent ads or being bombarded with more of them" has become a major pillar of the global "fraud economy."
These previously unpublished documents also reveal that the social media giant failed to identify and block a massive amount of advertising for at least three years, exposing billions of users on its social media platforms, Facebook, Instagram, and WhatsApp, to fraudulent e-commerce and investment scams, illegal online casinos, and the promotion of prohibited medical products.
A document from December 2024 indicated that Meta's platforms displayed an average of approximately 15 billion "high-risk" fraudulent ads to users every day—ads that clearly exhibit fraudulent behavior. Another document from late 2024 showed that Meta earned approximately $7 billion (about 50 billion yuan) annually from these fraudulent ads.
The documents show that most fraudulent activities stemmed from suspicious actions by advertisers, actions that should have triggered Meta's internal alert system. However, the company only blocked marketers when its automated system predicted a 95% or higher probability of fraud. If the system determined a lower probability but still suspected the advertiser of fraud, Meta would penalize them by increasing their ad rates, a move aimed at preventing suspicious advertisers from running ads.
The document further points out that because Meta's ad personalization system pushes ads based on user interests, users who click on fraudulent ads are likely to see more of these ads.
The documents also reveal that internal Meta research indicates its products have become a significant pillar of the global "fraud economy." A May 2025 presentation by its security team estimated that the company's platform was involved in one-third of all fraud cases in the United States. Meta has also acknowledged in other internal documents that some of its major competitors are doing a better job of cleaning up fraudulent activity on their platforms.
An internal assessment report from April 2025, focusing on platforms where scammers operate, concluded that "it is easier to run fraudulent ads on Meta than on Google." The document did not explain the basis for this conclusion.
Meta responds: The actual percentage is lower; the SEC is investigating.
In response to the media reports, Meta spokesperson Andy Stone stated that the documents seen by the media "present a selective perspective and distort Meta's strategy for dealing with fraud and scams."
He pointed out that the company's internal estimate that 10.1% of revenue in 2024 would come from fraudulent and other illegal advertising was "too rough and too broad." He explained that subsequent investigations by the company revealed the actual percentage to be lower, as the estimate included a "large amount" of legitimate advertising. However, he declined to disclose the latest figures.
“This assessment was conducted to verify our planned investments in integrity—including investments in combating fraud and scams—and we have indeed done so,” Stone added. “We actively combat fraud and scams because people on our platform don’t want this content, legitimate advertisers don’t want it, and we don’t want it either.”
The disclosure of these documents comes at a time when global regulators are pushing Meta to take further steps to protect users from online fraud. According to internal documents, U.S. Securities and Exchange Commission... The SEC is investigating Meta for publishing fraudulent financial advertisements.
Last year, UK regulators also revealed that 54% of losses in payment fraud in 2023 were related to Meta products, more than twice the total of all other social media platforms combined.
Net profit plummeted 83%, stock price plunged
In late October, Meta released its Q3 2025 financial report. Although revenue increased by 26% year-over-year, a sharp 83% drop in net profit caused its stock price to plummet by 11.33%, marking its biggest single-day drop in three years and wiping out $214.7 billion in market capitalization. This is also Meta's second-largest single-day market capitalization loss in history, second only to February 3, 2022.
Meta's revenue reached $51.24 billion in the third quarter, a 26% year-over-year increase, indicating that its core advertising business continues to grow steadily.
The company's net profit plummeted to $2.71 billion for the quarter, with diluted earnings per share of only $1.05, primarily due to a one-time non-cash income tax charge of $15.9 billion imposed by the new U.S. tax law.
Meta's CFO, Susan Lee, explained in the earnings call that this was a one-time accounting treatment that "does not involve any actual cash outflow" and will bring significant cash tax savings to the company in the future. Excluding income tax, Meta's effective tax rate in the third quarter was approximately 14%, and its adjusted net profit would reach $18.6 billion.
Financial reports show that the AI arms race among tech giants has entered a fierce stage. Meta continues to invest heavily in AI. The company announced that its capital expenditures in 2025 are expected to reach $70 billion to $72 billion, an upward revision from the previous forecast of $66 billion to $72 billion, and expects total spending growth in 2026 to "significantly exceed that of 2025," mainly driven by the construction of AI infrastructure. Wall Street analysts generally believe that Meta's spending in this area next year will jump from "no more than $72 billion" this year to "$97 billion." Zuckerberg clearly stated in the earnings call that the company will "aggressively" increase spending on artificial intelligence. Maintain competitiveness in the competition.
Meta's AI strategy focuses on two main directions: first, developing next-generation general-purpose AI models through the Meta Super Intelligence Lab (MSL); and second, promoting the implementation of AI hardware products, such as the best-selling Ray-Ban Meta smart glasses.
Meta is also actively building up its resources. A filing with the U.S. Securities and Exchange Commission (SEC) on October 30th revealed that Meta Platforms plans to raise up to $30 billion in senior notes to support its artificial intelligence and data center initiatives. Investment. Meta's bond issuance attracted approximately $125 billion in investor subscriptions, setting a new record for the largest public corporate bond issuance in history (the previous record being $120 billion set by CVS Health in 2018 for its acquisition of Aetna Life). Wall Street analysts believe that the five-fold oversubscription indicates institutional investors' strong support for Meta's long-term strategy in artificial intelligence . This is also Meta's first bond issuance since its initial foray into the bond market in 2022, when the company issued $10 billion in bonds.
(Source: Daily Economic News)
