In case of synthetic fraud, the victim remains oblivious of the deceit unless made aware

DR. ARINDAM SARKAR
HOD AND ASSISTANT PROFESSOR, DEPARTMENT OF COMPUTER SCIENCE & ELECTRONICS-RAMAKRISHNA MISSION VIDYAMANDIRA, BELUR MATH

“Talking about synthetic fraud, we understand that forces are trying to create fictitious identities and are trying to fabricate the data. They are trying to blend real and fake personal information. The problem is that no real person is affected directly. That is why victims stay unaware, and we do not have the technology to check whether a video is real or fake. All of us are aware that we have used our VID (Virtual ID) to save our Aadhaar information. I can break the VID and can gather all your information. This information is actually being taken for data connection purposes. After collecting the data, I can fabricate the identity and then apply for the loan. I can build credibility and then after taking the loan, I will default.

So what are the different types of synthetic frauds? It starts with credit card fraud, where the fraudster will build the credit history and then default on a large amount. Then there is employment fraud, where the fraudster will create fake identities for jobs and trick a person looking for a job into believing it. Likewise, there are several different frauds which includes government benefit frauds for pension account holders insurance or healthcare frauds, or the SIM card fraud in the telecom industry where fake identities are created for issuing dubious SIM cards. The challenge behind these frauds is the inability to detect the real fraudster or the victim. There is also the problem of data silos which means we are not able to gather all the data together. Suppose you know the machine learning and deep learning algorithm and the fuel for this is the data. So if you collaborate all the data together, only then you will be able to proceed to train your machine or deep learning algorithm. But the problem is– Naqli Chehra, Asli Khel, Suraksha ka ho gaya Fail!

So how can we tackle this whole problem of deep fake? The first is to detect the Identification Method; second is doing Psychological or Physiological Liveness check; and the third is Audio Analysis. So presenting before you the Faceoff – a Revolutionary AI Platform for Digital Authenticity and Deepfake Detection. Under this platform we have eight different modules. Faceoff addresses the challenges of deepfake attacks, digital fraud, and authenticity challenges through advanced multimodal AI capabilities that analyze multiple behavioral and biometric indicators simultaneously, delivering unprecedented accuracy in digital authenticity verification.”