日本語
한국어
Tiếng Việt
简体中文
A recent investigation, based on extensive internal documents, has brought to light how Meta has allegedly leveraged billions from scam advertisements, a substantial portion of which was earmarked to fuel its ambitious artificial intelligence (AI) growth.
The bombshell report by Reuters exposes a five-year period of Meta’s practices on platforms like Facebook, Instagram, and WhatsApp, detailing how the tech giant projected significant earnings by tolerating and even targeting users most susceptible to fraudulent ads.
High Risk, High Value: Meta’s Approach to Scams
Internal communications reveal Meta’s hesitation to promptly remove accounts, even those identified as operating “scammiest” schemes. This reluctance reportedly stemmed from concerns that an abrupt drop in advertising revenue could divert crucial resources from Meta’s burgeoning AI initiatives.
Instead of swift action, Meta allowed “high value accounts” — often associated with bad actors — to accumulate “more than 500 strikes” without immediate suspension. Intriguingly, Meta’s documents indicate a strategy where these scammers were “penalized” by being charged higher ad rates. Simultaneously, Meta acknowledged its ad-personalization systems inadvertently aided scammers by delivering these fraudulent ads to users most likely to click on them, effectively trapping vulnerable individuals in a cycle of exposure.
Meta’s internal estimates paint a stark picture: users across its platforms encounter an astonishing 15 billion “high risk” scam ads daily, in addition to 22 billion organic scam attempts. Projections for last year suggested that approximately $16 billion, representing about 10 percent of Meta’s total revenue, would originate from scam advertisements.
Types of Scams and Their Impact
“High risk” scam ads encompass a wide array of deceptive schemes, from selling counterfeit products and fraudulent investment opportunities to promoting banned medical items or illegal online casinos. Of particular concern to Meta are “imposter” ads, which mimic celebrities or major brands. The company feared these imposter scams could prompt legitimate advertisers to withdraw from its platforms, impacting brand engagement and advertising revenue.
For example, scam ads featuring images of prominent figures like Elon Musk or Donald Trump, promising gifts or tariff relief, were flagged by Reuters. Another particularly insidious scam posed as a legitimate law firm offering advice on avoiding online fraud. While Meta removed these specific examples after being notified, the company still reportedly earned about $7 billion from similar “high risk” ads in 2024 alone.
Calls for Regulatory Intervention and Meta’s Defense
Sandeep Abraham, a former Meta safety investigator and current fraud examiner, emphasized the need for regulatory intervention. “If regulators wouldn’t tolerate banks profiting from fraud, they shouldn’t tolerate it in tech,” Abraham stated, highlighting the ethical implications of profiting from deception.
Meta spokesperson Andy Stone refuted the report’s conclusions, claiming the presented documents offer “a selective view that distorts Meta’s approach to fraud and scams.” Stone suggested Meta’s $16 billion revenue estimate from scam ads was “rough and overly-inclusive” and that the actual figure was much lower, though he declined to provide an exact amount. He affirmed Meta “aggressively fight[s] fraud and scams” because such content is unwanted by users, legitimate advertisers, and the company itself.
Despite these assurances, internal Meta documents from earlier this year revealed the company’s platforms were implicated in a third of all successful scams in the U.S. Furthermore, staff concluded it was “easier to advertise scams on Meta platforms than Google,” acknowledging rivals’ superior fraud detection capabilities.
Enforcement Challenges and Transparency Debates
Meta claims significant improvements, citing a 58 percent reduction in user reports of scam ads globally and the removal of over 134 million pieces of scam ad content so far in 2025. However, the report raises questions about the pace and prioritization of Meta’s enforcement efforts.
In 2023, Meta reportedly laid off its entire team dedicated to advertiser brand-rights issues and instructed safety staffers to reduce computing resource usage, shifting focus towards virtual reality and AI. A 2024 document suggested a “moderate” enforcement approach, aiming to incrementally reduce revenue from scams, illegal gambling, and prohibited goods by 1-3 percentage points annually, with a goal of halving it by 2027. More recent 2025 documents indicate ongoing internal discussions about how “abrupt reductions of scam advertising revenue could affect its business projections.”
Compounding these concerns, a team responsible for vetting questionable advertisers was reportedly told not to take actions that could cost Meta more than 0.15 percent of its total revenue – roughly $135 million for any single scam account. While Stone denied a “hard limit,” the manager’s internal warning to “be cautious” about losing revenue by mistakenly blocking “benign” ads underscores the financial considerations at play.
Documents also suggest Meta prioritized action primarily when faced with regulatory fines, even though potential revenue from scam ads far outweighed the highest possible penalties. Police requests for scam enforcement, such as 146 examples from Singapore, often saw limited action, with only 23 percent violating Meta’s policies and the rest merely violating “the spirit of the policy, but not the letter.”
Rob Leathern, who formerly led Meta’s business integrity unit, and Rob Goldman, Meta’s former vice president of ads, have launched CollectiveMetrics.org, a nonprofit advocating for greater transparency in digital advertising. They argue that access to large, random samples of ads for independent researchers is crucial to objectively assess how effectively platforms combat fraud. Leathern also recommends that platforms like Meta should not only notify users who have clicked on scam ads but also donate ill-gotten gains to nonprofits dedicated to educating the public about recognizing online scams.