India is the largest democracy in the world. Our country conducts the world’s biggest elections with great pride and accountability. Voter turnout hovers around two-thirds despite massive expenses and logistics. So, how can we get everyone to vote? Let us think from a sociological perspective. We are investing so much through what we call special intensive revision (SIR) by the Election Commission (EC) of India to get all eligible voters registered and verified. This must bring us the desired dividends to strengthen democracy. Our election system is robust, but there are some issues like limited end-to-end verifiability, duplicate and/or impersonation risks, and access gaps for migrants, the elderly, and persons with health challenges. When addressed, these will make our election governance system technologically up-to-date and robust as well as socially inclusive.

Drawing from one of our concurrent research studies, in this discourse, we argue that current EVM+VVPAT processes do not give every voter a way to personally confirm their ballot on a public record, and manual ID checks err, if any. An online voting option, done right, could help close these gaps and strengthen our democratic process while reducing expenses. Our proposal can be understood in four simple steps, which include registration, selfie check, vote /re-vote, and verification with a scannable slip. First, it requires voters to register and sign in, where voters can use standard credentials such as EPIC or OTP as permits to proceed. In this, minimal personal data touches the blockchain ledger. The design keeps real identities off the ledger entirely. Second, we call for a live selfie check. This can be at the booth or can be done remotely. The system runs an AI verification of the live selfie to confirm “you” are you. Interesting but very pertinent. Then it performs a fast lookup to ensure you have not already cast a valid ballot elsewhere. Here, no raw face images are stored and only encrypted numerical templates sit off-chain. Third, we actualise what a vote or re-vote is. Herein, you select your candidate and submit an encrypted ballot to a permissioned shared ledger. This ledger can be jointly watched by the election authority, contesting parties, and members of civil society. If you want to re-vote at a supervised booth after voting at a remote location, you can do so, and the system will keep only your latest vote for tallying. Finally, we verify privately. You get a QR receipt—a scannable slip—to confirm your ballot is on the ledger and for which candidate- without disclosing your identity to others. It contains a one-time secret that lets only you check your own encrypted ballot entry.

Accessibility, safety, and privacy are the hallmarks of this proposed online voting. The design draws a hard line- no biometrics on the ledger. The ledger stores only pseudonyms. Encrypted ballots, personal details, and facial templates stay off-chain under encryption. Multiple trustees like election authorities, contesting parties, and members of civil society hold decryption keys, as no single actor can read vote ballots or change results. The QR slip helps the voter verify inclusion and choice without revealing one’s identity to anyone else. Every update leaves an audit trail of versions and timestamps. The selfie check itself has guardrails. Liveness checks—like a blink or slight head turn—run server-side to defeat printed photos, replays or deepfakes. A “one-face only” rule blocks supervised coercion attempts where another face hovers in frame.

The whole process is coercion resistance for the public, and the latest vote wins. Let us explain with an example. Suppose someone pressures Radha to vote a certain way. She can later go to a supervised booth to re-vote, or she can re-vote in private. The ledger keeps both entries for audit, but only her latest vote will count in the tally. The QR receipt Radha holds allows her to confirm that her final encrypted ballot is the one recorded. This reduces the payoff from coercion and vote-buying because a coerced vote can be overwritten. This is extremely useful for governance and scalability. This is not a “single server” plan. It can be a consortium model- a permissioned blockchain ledger which, when run by the Election Commission, can be under the observation of all contesting parties as well as of civil society members. It will spread trust and create an audit trail. A test prototype focuses on the AI verification path and shows the core “selfie check + fast lookup” running in milliseconds. In short: seconds, not minutes, for the voter experience at scale.

There are some limitations, as it requires investment—assisted voting at volunteer booths for people without devices or with low digital literacy. And the legal process—rules for online voting, data retention and privacy—must be clearly defined before any rollout. Nevertheless, these limitations, when seen in the light of benefits, are minuscule. Below, we have open questions and their brief answers to remove any doubt about the technical jargon or engineering behind the quest for online voting and for better comprehension of this. Please know that online voting is going to be the order of the day tomorrow. Sociological imagination suggests that the sooner we switch to online voting in India, the better it will be for the public at large and for our election governance.

Here are some jargon busters and assumed frequently asked questions (FAQ) –

***

Aditya Raj is an Associate Professor at the Department of Humanities and Social Sciences, Indian Institute of Technology Patna. He completed his PhD as a Commonwealth Fellow from Canada’s McGill University in 2008 after which he was at the University of British Columbia before joining IIT Patna in 2010. Himanhsu Sharma is a Director with the Department of Telecommunications, Government of India, with 16+ years of experience in ICT, digital infrastructure, and public-sector technology governance. He holds an M. Tech in Data Science from BITS Pilani and is pursuing a PhD at IIT Patna.