Meta Cracks Down on Deepfake Scams Targeting Brazil and India

In an aggressive move to curb a rising tide of online fraud, Meta has dismantled over 23,000 Facebook pages and accounts tied to investment and payment scams, primarily targeting users in Brazil and India. These scams leveraged advanced deepfake technology, creating realistic but entirely fabricated videos of personal finance influencers, cricket icons, and prominent business personalities endorsing bogus financial platforms.

The fraudsters used these impersonations to lure unsuspecting users toward fraudulent investment apps and gambling websites. Some of the most alarming schemes involved fake Google Play Store replicas that tricked users into downloading malware-laden apps. Many of these campaigns redirected victims to messaging platforms like WhatsApp, where scammers posed as financial advisors pushing false investment tips.

This large-scale takedown is part of

A Breakdown of the Scam Landscape

Here’s a condensed overview of the types of fraud Meta is fighting:

Investment Scams: Often masquerading as financial experts, scammers promise rapid, high returns from crypto, stocks, or real estate. Deepfake videos make their claims seem more credible.

Advance Payment Scams: Typically seen on Facebook Marketplace, fraudsters list items for sale and demand upfront payments—then vanish without delivering the product.

Overpayment & Refund Scams: A fake buyer overpays using a forged receipt, asks for a refund of the difference, and later reverses the original transaction—leaving the seller out of pocket.

Meta’s India-Focused Anti-Scam Efforts

Meta has deepened its partnership with Indian authorities, emphasizing education and enforcement:

With the Department of Telecommunications (DoT): WhatsApp hosted workshops to train officials in identifying and reporting scams effectively.

With the Department of Consumer Affairs (DoCA): A co-led digital literacy campaign under Jago Grahak Jago educated Indian citizens about fraud indicators.

With the Indian Cybercrime Coordination Centre (I4C): Meta conducted law enforcement training across seven states to empower cybercrime units in handling online fraud cases.

Platform Tools Built for User Protection

Meta is integrating multiple layers of protection across its platforms:

Messenger Warnings: These in-app alerts flag suspicious payment-related messages or patterns tied to scams.

Selfie Verification: A biometric feature used to verify real users, counter impersonation attempts, and recover hijacked accounts.

Privacy Check-Up: An interactive tool that helps users limit who can view their profiles, message them, or access sensitive details.

What Undercode Say:

Meta’s latest crackdown reflects not just a reactionary stance but a calculated pivot toward preemptive platform security. The sheer scale of deepfake use in these scams points to the sophistication of digital fraud in 2025. Instead of rudimentary phishing attempts, we’re now seeing AI-generated faces and voices used to create high-conviction traps for users seeking quick financial gains.

This pattern represents a concerning fusion of misinformation, AI abuse, and financial exploitation. Platforms like Facebook, with their massive reach in emerging markets like India and Brazil, are prime battlegrounds for this new breed of cybercriminals. Deepfakes featuring familiar public figures make the scams feel local, personal, and therefore more persuasive.

Meta’s collaboration with Indian authorities signifies an important strategy: decentralizing the fight against scams by engaging not just tech teams, but telecom agencies, consumer protection departments, and law enforcement. These partnerships allow a multi-pronged response that goes beyond platform moderation.

But questions remain. For instance, will the selfie verification system prove robust enough in high-volume countries like India? How does Meta plan to scale content moderation when deepfake production becomes even more automated? The reliance on facial recognition also introduces privacy concerns in markets already wary of surveillance technologies.

Undercode believes the answer lies not just in moderation but in proactive friction—introducing deliberate hurdles in risky interactions, such as financial transactions with unknown parties or new app installations via third-party links. A user should never be one click away from downloading a scam.

Moreover, user education campaigns must evolve. Instead of reactive alerts after an action, platforms should prioritize predictive nudges—contextual reminders when users are about to engage in potentially unsafe behavior. Coupling these UX strategies with AI-powered detection systems could provide the layered defense today’s digital ecosystem demands.

As AI-generated content becomes harder to distinguish from real media, the line between fiction and fraud continues to blur. Meta’s success won’t be measured by account takedowns alone, but by how well it restores user trust through transparency, safety, and collaboration.

Fact Checker Results

Meta has confirmed the takedown of over 23,000 accounts. The deepfake scam techniques and targeting of India and Brazil have been substantiated by Meta’s public safety updates and security transparency reports. Collaborations with Indian government bodies are active and ongoing.

Prediction

Expect deepfake scams to become more interactive, with real-time AI-driven conversations imitating trusted voices. The future of online fraud will rely less on static content and more on AI-powered engagement. Platforms will need to shift from static content moderation to dynamic behavioral analysis, recognizing scam patterns as they unfold in real time.