Internal documents reviewed by Reuters describe a difficult and damaging tension inside Meta: scam ads were creating user harm, but removing them too quickly could also hit revenue that the company was counting on for growth, including artificial intelligence.
The documents, created between 2021 and 2025 by Meta’s finance, lobbying, engineering, and safety divisions, show how fraud on Facebook, Instagram, and WhatsApp became more than a moderation problem. It became a business issue, an enforcement issue, and a trust issue at the same time.
What the documents say
Reuters reported that Meta internally projected billions of dollars in revenue from scam ads. The documents showed that Meta’s systems could help deliver scam ads to users most likely to engage with them, because ad personalization is built around inferred interests and past behavior.
That created a troubling loop. A user who clicked on scam ads could be shown more of them, because the platform interpreted the interaction as a signal of interest. Reuters summarized the issue this way: “Users who click on scam ads are likely to see more of them because of Meta’s ad-personalization system, which tries to deliver ads based on a user’s interests.”
Meta’s internal estimates were large. Users across its apps were estimated to encounter 15 billion “high risk” scam ads a day, along with 22 billion organic scam attempts daily, according to a 2024 document cited in the report. Last year, Meta projected that about $16 billion, or about 10 percent of its revenue, would come from scam ads.
The documents also described uneven enforcement. Reuters reported that some “high value accounts” were allowed to “accrue more than 500 strikes without Meta shutting them down.” Instead of immediate removal, Meta documents showed that the company “penalized” scammers by charging higher ad rates as strikes accumulated.
Why enforcement was complicated
The source material describes Meta as hesitant to remove some bad actors abruptly, even those internally regarded as among the “scammiest scammers.” The reason was not presented as a technical limitation alone. Documents showed concern that a sudden revenue decline could reduce resources needed for artificial intelligence growth.
Reuters also reported that Meta took a “moderate” approach to enforcement in a 2024 document, aiming to reduce revenue “attributable to scams, illegal gambling and prohibited goods” by 1–3 percentage points each year since 2024, with the goal of cutting it in half by 2027. A 2025 document showed Meta still weighing how “abrupt reductions of scam advertising revenue could affect its business projections.”
The balance became sharper as Meta needed substantial resources for AI. Reuters reported that Meta required $72 billion to invest in AI, while also taking steps that would limit the revenue impact of scam-ad enforcement.
In February, Reuters reported, Meta told “the team responsible for vetting questionable advertisers” that they were not “allowed to take actions that could cost Meta more than 0.15 percent of the company’s total revenue.” Reuters noted that this amounted to any scam account worth about $135 million. Meta spokesperson Andy Stone disputed that interpretation, saying the team was never given “a hard limit” on what the manager called “specific revenue guardrails.”
The kinds of scams involved
The “high risk” scam ads described in the report included fake products, investment schemes, banned medical products, illegal online casinos, and ads from sketchy entities. Meta was especially concerned about “imposter” ads that copied celebrities or major brands, because those scams could push advertisers or users away from its apps.
Reuters identified examples that used well-known figures. One ad using Elon Musk’s photo read, “Hey it’s me,” and “I have a gift for you text me.” Another ad using Donald Trump’s photo claimed the US president was offering $710 to every American as “tariff relief.” A third posed as a real law firm and offered advice on avoiding online scams.
Meta removed those ads after Reuters flagged them. But Reuters reported that in 2024, Meta earned about $7 billion from “high risk” ads like those alone.
Other scam categories described in the documents included crypto scams, fake concert tickets, fake job ads claiming to hire for Big Tech companies, and promotions that looked “too good to be true,” including 80 percent off a desirable item from a high-fashion brand.
Meta’s response
Andy Stone told Reuters that the documents “present a selective view that distorts Meta’s approach to fraud and scams.” He said the estimate that 10 percent of 2024 revenue would come from scam ads was “rough and overly-inclusive,” and he suggested the real amount was much lower, while declining to provide a figure.
Stone also said Meta’s investor disclosures note that scam ads “adversely affect” revenue. He defended the company’s position directly: “We aggressively fight fraud and scams because people on our platforms don’t want this content, legitimate advertisers don’t want it, and we don’t want it either.”
Meta also pointed to improvements. Stone told Reuters that “over the past 18 months, we have reduced user reports of scam ads globally by 58 percent and, so far in 2025, we’ve removed more than 134 million pieces of scam ad content.”
At the same time, Reuters reported that Meta’s own safety team estimated this spring that the company’s platforms were involved in a third of all successful scams in the US. Around the same period, internal documents said “it is easier to advertise scams on Meta platforms than Google,” while acknowledging that rivals were better at “weeding out fraud.”
The larger trust problem
The documents suggest that the core issue is not only whether Meta can detect scams. It is whether the company moves quickly enough when enforcement affects revenue, advertisers, computing resources, and business projections.
Reuters reported that in 2023, Meta laid off “everyone who worked on the team handling advertiser concerns about brand-rights issues,” and later told safety staffers to limit computing resources so more could be devoted to virtual reality and AI. Stone said Meta eventually “substantially expanded” its teams tracking scam ads.
Former Meta safety investigator Sandeep Abraham, now running consultancy firm Risky Business Solutions as a fraud examiner, told Reuters that regulators should step in. His argument was blunt: “If regulators wouldn’t tolerate banks profiting from fraud, they shouldn’t tolerate it in tech.”
Rob Leathern, who previously led Meta’s business integrity unit and left in 2020, told Wired that it is hard to “know how bad it’s gotten or what the current state is” because Meta and other social media platforms do not give outside researchers access to large random samples of ads.
That lack of visibility matters. Without broad outside access, the public has to rely on internal documents, company statements, regulator pressure, and reports from users or police. The Reuters findings show why that is not enough for many critics: when scam ads are also a revenue stream, every enforcement decision carries financial consequences.