Meta Under Fire for Failing to Tackle Surge in AI Scams
Meta faces mounting criticism for its handling of the surge in AI-driven scams. The company is under scrutiny as experts and consumer groups warn that insufficient action against fraudulent schemes is putting millions of users at risk.
Tl;dr
Deepfake Scams Surge Across Meta‘s Platforms
In recent months, a surge of sophisticated scams leveraging AI-generated deepfakes has posed a significant threat to Meta. The proliferation of accessible tools now enables bad actors to convincingly impersonate the images and voices of high-profile celebrities. From Cristiano Ronaldo to Mark Zuckerberg, and even actress Jamie Lee Curtis, fraudulent videos have become alarmingly common on Facebook. According to the independent Oversight Board, thousands of deceptive ads featuring such fakes are circulating, underlining the scale of the problem.
Ineffective Moderation: A System Under Strain
Despite numerous user reports, fraudulent content often remains visible for extended periods. Take the recent example: a fake testimonial video generated by AI using ex-footballer Ronaldo Nazário‘s likeness promoted a dubious online « Plinko » game. This ad, despite being flagged over fifty times for fraud, racked up more than 600,000 views before any action was taken—prompted only after intervention from the Oversight Board. The Board’s ruling highlighted a critical flaw: «Meta likely allows substantial fraudulent content to slip through in order not to hamper legitimate partnerships with celebrities.»
A closer look at the company’s moderation practices reveals deeper concerns. Moderators reportedly lack both the resources and training required to identify these new forms of impersonation, largely due to insufficient guidelines and vague internal instructions. As a result, inconsistencies continue to undermine efforts against such scams.
External Pressure and Corporate Inertia
Concerns about lax oversight are not limited to internal critics. As reported by the Wall Street Journal, nearly half of all Zelle-related fraud cases at JPMorgan Chase now originate from Meta’s platforms. Regulatory bodies in the UK and Australia echo similar warnings, signaling mounting global unease. Still, the company appears reluctant to impose tighter restrictions on advertisers—despite acknowledging that «criminal activity is increasing but so are our countermeasures; we are piloting facial recognition and additional tools for user protection.» Even so, Meta pushes back against its critics, asserting that «a number of the Board’s claims are simply inaccurate.»
The Call for Change: Awaiting Action
Against this backdrop, only one clear recommendation stands: the Oversight Board urges Meta to update its internal policies and better equip moderators to tackle AI-driven scams. While both regulators and users wait for an official response within sixty days, the expectation is clear—a meaningful shift in approach is now overdue.