The dark side of Meta When "scammer ads" are a gold mine that AI intentionally lets through?
The Big Tech world is reeling from a fresh scandal as former executives and employees of Meta (the parent company of Facebook and Instagram) have come forward with internal documents and evidence. The disclosure suggests that Meta’s vast advertising ecosystem may be prioritizing revenue generation over user safety, effectively turning a blind eye to a "scam economy" that impacts millions.
The 19% Revenue Revelation
Internal leaks indicate that Meta has long been aware of the massive revenue generated from "bad actors." According to the documents, an estimated 10-19% of total ad revenue originates from high-risk or illicit categories, including scams, online gambling, and counterfeit goods. The allegations suggest that some top-tier executives were aware of these figures but opted to "delay" aggressive enforcement actions to maintain strong quarterly earnings for investors.
"Surplus Pricing": Profiting from Deception
Perhaps the most shocking revelation is a mechanism allegedly called "Surplus Pricing." Instead of immediately banning advertisements flagged as potentially fraudulent, the system purportedly applies a premium rate to them. If the algorithm detects a high probability of a scam (yet below the 95% threshold for an automatic ban), it increases the ad cost. This allows Meta to extract higher profits from scammers while the scammers continue to reach potential victims as long as they pay the premium.
The Personalized Ad Loophole
Whistleblowers also warn that Meta’s sophisticated Personalized Ads AI is a double-edged sword. Once a user accidentally clicks on a deceptive ad, the AI categorizes them as a "high-quality lead" for that specific niche. Consequently, the system relentlessly targets that user with similar fraudulent content, creating a predatory cycle that is difficult for victims to escape.
Meta’s Defense
Meta has denied parts of these allegations, stating that the company invests billions annually into safety and security. They characterized the leaked revenue figures as "misleading estimates." Nevertheless, the global tech community and ethicists are raising critical questions about the morality of a business model that balances human harm against profit margins.
- In the online advertising industry, there's a term called "Arbitrage of Truth," or profiting from gray areas. This leaked information suggests that platforms may intentionally set excessively high ban thresholds to allow borderline ads to remain active for as long as possible before being removed.
- A 2025 report indicated that most scams on Meta originate from transnational networks using AI to create fake images and audio (deepfakes). Meta's "slow" detection system allows scammers to rake in massive amounts of money within hours before pages are shut down.
- This case could become crucial evidence in the US Congress to push for the "Algorithm Accountability Act," which would require social media companies to publicly disclose their "profit-making algorithms" to ensure they aren't profiting from crime.
- The alarming aspect is the use of the "Sunk Cost Fallacy," where victims are manipulated into believing they've already lost money and are forced to transfer more. The platform's AI has made it alarmingly accurate in identifying victims with these psychological traits.
Meta Rebrands Horizon Worlds to "Worlds" Pivoting to Mobile to Rival Roblox.
Source: The Verge
Comments
Post a Comment