Deepfake Investment Scam: Analysis of the Valueleaf Case

Deepfake powered financial fraud amplified by ad tech loopholes

The October 2025 arrest of four senior employees from the ad tech firm Valueleaf by the Mumbai Police marks a pivotal moment in India’s ongoing fight against AI enabled financial crime. This case exposed how deepfake technology, combined with structural weaknesses in the digital advertising ecosystem, can be weaponized to execute large scale financial fraud with alarming ease.

 

The Breach of Trust and Technology

Valueleaf Group, headquartered in Bangalore, is a well established MarTech and AdTech company known for delivering performance driven digital marketing solutions. Recently acquired by VerSe Innovation, the parent company of Dailyhunt and Josh, the firm expanded beyond India into the UAE and the United States.

As a certified Meta Business Partner, Valueleaf enjoyed a high trust score and a credit limit reportedly around Rs 130 crore on Meta’s advertising ecosystem. This enabled the company to run massive advertising campaigns on credit, without immediate billing, a level of access given only to a small number of agencies.

Valueleaf’s services include complete digital growth strategies for customer acquisition, engagement, retention, data driven marketing, SEO, mobile integration, and high performance advertising across social platforms. As an official Meta partner, it helps brands optimize campaigns using advanced analytics and AI powered marketing tools.

This reputation and privileged access became the perfect gateway for a foreign cyber fraud network to infiltrate India’s digital advertising infrastructure.

Valueleaf is reportedly serving major banking institutions across India, sources said.

 

Anatomy of a Coordinated Fraud

Between July 1 and July 18, 2025, a China based cyber fraud syndicate, allegedly operating behind a Hong Kong front company named First Bridge, carried out one of India’s most significant deepfake driven investment scams.

At the center of the operation were AI generated deepfake videos impersonating well known Indian stock analysts and business news anchors. These extremely realistic videos were pushed as paid advertisements on Facebook and Instagram, exploiting Meta’s trust based advertising layer.

Once victims clicked the ads, they were redirected to WhatsApp groups handled by overseas scammers posing as financial experts. Using psychological manipulation, bogus dashboards, and fabricated profit screenshots, victims were convinced to deposit money into mule bank accounts. When they attempted to withdraw funds, they were blocked, pressured for additional payments, or simply ignored. The fraudsters then vanished with the money.

The scam surfaced after SEBI registered research analyst Prakash Gaba, age 73, filed a complaint leading to an FIR on September 29. Investigators discovered that one of the advertisement accounts used in the scam belonged to a woman from Assam who had left Valueleaf in 2024, yet her account access was still active.

Senior Police Inspector Suvarna Shinde stated that the four accused employees knowingly provided advertising access to the fraudsters for Rs 3 crore and continued enabling the advertisements even after realizing they were fraudulent.

 

Who Paid for the Campaign

The entity that contacted Valueleaf claimed to be a Hong Kong based company. Police investigations later revealed:

• The backend infrastructure was actually operated from Vietnam
• Financial and account patterns matched earlier China linked cyber fraud networks

This method is familiar in cross border scams:

1. Register a clean Hong Kong company
2. Run the scam system from Vietnam or mainland China
3. Use trusted Indian agencies to lend legitimacy

Despite the fraud, the ultimate revenue beneficiary was Meta, which collected earnings from every rupee spent on the paid ads.

Deputy Commissioner of Police Purushottam Karad explained that Meta flagged the advertisements as suspicious after seven days. In response, the accused expanded their operations from 18 advertisement accounts to 38 in an attempt to outrun detection. Meta eventually shut down these accounts on July 18, after which the fraudsters continued the campaign through a company in Dubai. The operation also used the internal code phrase “No Eye Gaming Black Hat Campaign.”

 

How the Scam Worked

Once the deepfake videos appeared as advertisements:

1. Users watched videos where Gaba appeared to endorse a fake investment scheme
2. Clicking the ads led them to WhatsApp groups run by foreign handlers
3. Fraudsters displayed manipulated profit screenshots and artificial dashboards
4. Victims deposited money into unrelated bank accounts
5. Withdrawal attempts were blocked or tied to additional fees
6. The scammers extracted all funds and disappeared

Thousands of Indian retail investors suffered financial losses estimated in crores.

Why This Scam Was Different

This represents India’s first major investment fraud involving complete AI generated deepfake videos, rather than earlier voice clone or chat based methods.

It relied on:

a) Highly realistic deepfake content

State of the art AI generated videos are increasingly indistinguishable from genuine footage.

b) Meta’s trust based advertising framework

Valueleaf’s trusted advertising identification allowed the scam to bypass automated checks.

c) Cross border syndicate operations

Vietnam and China based networks used a new model:

• A legitimate Indian Meta partner
• High trust advertising accounts
• High quality deepfake content
• Paid amplification at scale

The Four Valueleaf Executives

Those arrested were:

• Gigil Sebastian
• Deepayan Banerjee
• Daniel Arumugham
• Chandrashekhar Naik

They allegedly:

• Provided Meta advertisement identification access to the foreign syndicate
• Allowed international scammers to run campaigns through trusted Indian accounts
• Expanded the advertising footprint deliberately to evade Meta’s checks
• Received Rs 1.64 crore for their role

Their actions allowed the scam to appear legitimate and avoid scrutiny.

The Scam’s Execution and Evasion

Victims were directed to fake trading platforms and WhatsApp groups where initial profits were fabricated to build trust. The fraud became clear only when withdrawals failed.

Even after Meta flagged the content, the employees reportedly continued cooperating, expanding operations from 18 to 38 accounts to exploit loopholes in Meta’s enforcement system.

Legal and Financial Ramifications

Police filed cases under various sections of the Bharatiya Nyaya Sanhita and the Information Technology Act of 2000, including charges related to cheating, impersonation, and forgery. Illicit gains from India alone are estimated at Rs 3 crore, with additional sums likely transferred abroad.

Authorities believe this is India’s first large scale deepfake investment fraud.

Systemic Vulnerabilities Exposed

The involvement of a reputable ad tech firm reveals major weaknesses in the digital advertising supply chain.

Major gaps include:

• Poor client verification and onboarding for foreign companies using Indian ad accounts
• Excessively lenient review for trusted partner accounts
• Weak real time detection of deepfake content in paid promotions
• Over Dependence on automated systems that lag behind the rapid progress of AI generation models

This case demonstrates how deepfake technology threatens public trust in digital finance and digital advertising.

 

Can Meta Detect Deepfakes? The Difficult Question

A crucial question arises:

If deepfake videos are being uploaded, why can Meta not detect and prevent them? Is the problem inability or a lack of willingness?

Meta claims to use:

• Artificial intelligence detection
• Video and image forensic analysis
• Authentication signals
• Behavioral monitoring
• Manual ad review teams

Yet deepfake technology improves far faster than these safeguards.

Experts note:

• Only about 30 percent to 40 percent of deepfakes are caught automatically
• Trusted business partner accounts receive lighter review
• The advertising ecosystem prioritizes speed and revenue over manual screening

This is the exact weakness exploited in the Valueleaf case.

 

Multi Layer Ad Agency Chains: The Real Problem

The fraud pathway looked like this:

China based syndicate → Hong Kong front company → Valueleaf advertisement identification → Meta platform

Meta saw:

• A trusted Indian partner
• Regular payments
• A clean record

So the system approved the advertisements.

This creates a cartel like network in the advertising ecosystem:

1. Meta provides the infrastructure and earns revenue
2. Agencies provide trusted identification and earn commissions
3. Fraudsters exploit both
4. Victims lose everything

In this structure, every rupee spent on fraudulent ads still benefits Meta. Strict enforcement would affect large volumes of legitimate advertising spend, so the system often remains lenient in practice.

 

Police Advisory and The Road Ahead

DCP Karad urged citizens:
“We appeal to people not to fall for such online advertisements. If anyone notices such content, please contact 1930 or visit the nearest police station.”

He added that this is the first share market scam gang of this kind to be busted in the country, and the breakthrough will help uncover other similar scams.

 

Conclusion: A Systemic and Possibly Strategic Failure

The Valueleaf case is more than a technological shortcoming. It represents a systemic failure involving:

• Weak deepfake detection
• A flawed trust based partner model
• Insufficient client verification
• Regulatory blind spots
• Ease of exploitation by cross border syndicates

As AI generated deception becomes more lifelike, India must strengthen regulation, platforms must implement stronger safeguards, and agencies must uphold ethical responsibility. Without this, public trust in digital systems will continue to deteriorate.