Google Gemini Misused in AI Deepfake Loan Scam: Ahmedabad Police Expose Aadhaar OTP Bypass

The darker side of artificial intelligence is no longer a distant threat—it’s showing up in everyday crime reports. What once required hacking skills and insider access can now be done with AI tools and a bit of planning. A fresh case from Ahmedabad shows just how easily systems we trust can be manipulated.

Ahmedabad: AI-powered deepfake scam bypasses Aadhaar safeguards

In a troubling incident, the Ahmedabad Cyber Crime Branch has arrested four individuals accused of using AI tools, including Google Gemini, to carry out a sophisticated loan fraud. The accused allegedly created deepfake videos to bypass identity verification systems linked to Aadhaar and diverted OTPs to execute unauthorized financial transactions.

The case came to light when a local businessman involved in import-export noticed something unusual-he had stopped receiving OTP messages from his bank for nearly two days. Realising that this was not a technical glitch, he approached the police.

During the investigation, officials uncovered that the mobile number linked to his Aadhaar had been changed without the usual OTP verification process. Even more concerning, his biometric data appeared to have been manipulated, allowing the fraudsters to gain unauthorized access.

Police found that the accused had opened a bank account in the victim’s name and taken a loan of Rs 25,000. They also accessed his DigiLocker account to retrieve personal documents, which were then used to strengthen their fraudulent applications.

Deepfake videos used to trick KYC systems

According to investigators, the gang created AI-generated deepfake videos of the victim to bypass Aadhaar-based e-KYC checks. These videos helped them impersonate the victim convincingly enough to fool verification systems. Once access was gained, they changed the registered mobile number, ensuring that OTPs were redirected to their own devices.

Using these methods, the accused attempted to open accounts across multiple banks. They successfully created one with Jio Payments Bank, through which they processed the loan.

Police also revealed that one of the accused was employed at a Common Service Centre (CSC), giving him access to Aadhaar-related systems. This insider access appears to have played a key role in executing the fraud.

The four arrested individuals have been identified as Kanubhai Parmar, Ashish Vanand, Mohammad Kaif Patel, and Deep Gupta.

A warning sign for digital India

This case highlights a serious gap between fast-growing AI capabilities and existing security systems. While platforms like Aadhaar and DigiLocker are designed to make life easier, they are now becoming targets for highly advanced fraud techniques.

For users, the takeaway is simple but important: if OTPs suddenly stop or accounts behave strangely, act immediately. For authorities and tech companies, the challenge is bigger - security systems must evolve as fast as the technology being used to break them.

Read More: OpenAI Is Building an AI Phone. Will It Take on the iPhone?