Indian Institute of Technology (IIT), Ropar, has prepared a digital tool named “FakeBuster” to identify and crack down on virtual imposters either creating a nuisance during online conferences, webinars, examinations, meetings, chats and socialising among others or even committing frauds entailing financial implications. The makers of “FakeBuster” claim it to be one of the first tools across the world to detect imposters using the “DeepFake” detection technology during live video conferencing.
In simple terms, this standalone solution enables a user (organiser) to detect if another person’s video is manipulated or spoofed during a video conferencing.
Imposters have been using real-time mimicked visuals (videos) known as “DeepFake” during meetings, webinars, online examinations and job interviews with a lot of adverse social, mental and financial implications. Further, through this mode, manipulated media contents are being used in spreading fake news, pornography and other such online content with major repercussions.
The platform is likely to benefit organisations — both the Government and private — as also individuals functioning from within the confines of closed rooms to run offices or events for more than a year now due to the Covid-19 pandemic.
“Sophisticated artificial intelligence techniques have spurred a dramatic increase in manipulation of media contents. Such techniques keep evolving and becoming more realistic. That makes detection difficult and it could have far-reaching security implications. Hence, a platform was conceived,” said Prof Abhinav Dhall from IIT Ropar, one of the four-member team that developed the “FakeBuster” in association with Monash University of Australia.
The virtual imposters were successful in their evil designs even in a meeting of the United Nations last year, and similarly, according to media reports, a virtual meeting of South African lawmakers was disrupted by hackers who flooded the video call with pornographic images. The virtual imposters also hurled racial and sexual insults in the meeting which were held by video conference calls due to lockdown restrictions last year.
A live meeting on Kashmir organised six months ago in which experts from several countries were participating was hacked with songs on Lord Ram and Hanuman playing in the background. The meeting had a group of Pakistani diplomats, along with some human rights “activists” and as the event progressed, pro-Indian accounts intervened the event and tried to stop Pakistani diplomats from lying on the Kashmir issue.
The tool was planned after cases of manipulations were reported in recent times from India and abroad which found their way into video-calling platforms through spoofing tools based on transfer of facial expressions.
These fake facial expressions are often convincing to human eye and can have serious implications.
The IIT claimed that the tool has achieved over 90 per cent accuracy ever since being used as pilot project in recent times. This software platform is independent of video conferencing solutions and has been tested with Zoom and Skype applications. The device has already been tested and would hit the market soon.
The Deepfake detection tool, “FakeBuster”, works in both online and offline modes. Since the device can presently be attached with laptops and desktops only “we are aiming to make the network smaller and lighter to enable it to run on mobile phones/devices as well”, informed Associate Professor Subramanian. He said the team is working on using the device to detect fake audios also.
Deeming a couple of platform unsafe, the Ministry of Home Affairs (MHA) had advised Government officials not to use a particular video meet app for holding meetings. According to the Government, the app can be easily hacked and confidential data and material stolen.
In its advisory for private individuals, the MHA issued detailed guidelines to prevent unauthorised entry in the conference room, to prevent an unauthorised participant to carry out malicious activity on the terminals of others and avoiding DOS attack by restricting users through passwords and access grant. A DOS (denial-of-service) attack is done by hackers to make a machine or network resource unavailable to its intended user (s).