Japanese regulators felt dismayed last year by a flood of obvious scam ads on Facebook and Instagram. These scams ranged from fraudulent investment schemes to fake celebrity product endorsements generated by artificial intelligence.
Article contents:
- Introduction: Fear of regulations
- The Ad Library trick
- The global playbook
- Verification cost and revenue
- Whack-a-mole: Moving the scam
- Meta’s official response
- Conclusion
Introduction: Fear of regulations
Meta, the social media giant, feared that Japan would soon force it to verify the identity of all its advertisers, internal documents reviewed by Reuters show. This step would likely reduce fraud but would also cost the company revenue. To fend off this threat, Meta launched an enforcement campaign to reduce the volume of violative ads. However, it also sought to make problematic ads less “discoverable” to Japanese regulators.
The Ad Library trick
Meta’s remedy relied on its “Ad Library,” a publicly searchable database. The company realized that Japanese regulators were searching it as a “simple test” of Meta’s effectiveness in tackling fraud. To perform better on this test, Meta employees found a way to manage what they called “perception of prevalence.” They identified keywords and celebrity names used by regulators, and then deleted seemingly fraudulent ads from the library.
Rather than telling an accurate story, Sandeep Abraham, a former Meta fraud investigator, said: “It’s now telling me a story about Meta trying to give itself a good grade to regulators.” The tactic succeeded in making search results look cleaner than they actually were.
The global playbook
Cleaning up search results was so successful that Meta, documents show, added the tactic to a “general global playbook” deployed against regulatory scrutiny in other markets, including the United States, Europe, and India. The playbook outlines Meta’s strategy to stall and delay advertiser verification unless new laws leave them no other choice.
Verification cost and revenue
The documents show that Meta decided against spending roughly $2 billion estimated for comprehensive verification, fearing the loss of up to 4.8% of its total revenue by blocking unverified advertisers. Instead, it decided to adopt a “reactive-only” stance—resisting regulation and accepting comprehensive verification only if lawmakers compelled it to do so.
Whack-a-mole: Moving the scam
Meta’s analytics showed that even when verification blocked ads in one market (such as Taiwan), those same ads continued to generate revenue for the company in other markets. If an unverified advertiser was banned in Taiwan, Meta would repeatedly display those ads to users elsewhere, creating a “whack-a-mole” dynamic where banned scam ads appearing in one jurisdiction pop up in another.
Meta’s official response
Meta spokesperson Andy Stone said that removing fraudulent ads from the library also means removing them from the platform, denying any deception. He stressed that the company set global goals to curb fraud and saw a 50% drop in user reports of scams over the past year.
Conclusion
These documents shed light on the ongoing tension between profitability and responsibility at major tech companies. While AI tools that make creating scams easier are flourishing, the need for rigorous and transparent verification measures remains more urgent than ever to protect global users.
Frequently asked questions
Q: Why does Meta resist verifying all advertisers?
A: Because it is expensive (around $2 billion) and could lead to significant revenue loss from small or unverified advertisers who make up a large portion of its income.
Q: What is the “playbook”?
A: It is an internal Meta strategy that includes steps to delay regulatory laws, offer temporary voluntary solutions, and clean up the platform’s image before regulators to avoid strict rules.
Q: Are scam ads really a big problem?
A: Yes, reports indicate that “high-risk” fraudulent ads can generate up to $7 billion in revenue for the company annually and cause heavy losses to users.
