The morning of May 3, 2026. Across examination centres throughout India, more than 22 lakh students had settled into their seats to write the NEET-UG examination. They carried with them two to three years of relentless preparation, the sacrifices their families had made, and a single foundational belief that the examination would be fair. Within days, the National Testing Agency (NTA) announced a decision that NEET-UG 2026 stood cancelled. A coaching institute operator in Latur was suspected of having accessed the question paper a full 10 days before the examination. A student in Kerala had sent a leaked PDF to his father the very night before. WhatsApp and Telegram had done the rest.This is not a new story. In 2024, NEET-UG was similarly compromised. Before that, there were paper leaks in multiple state-level examinations. After each one, the same sequence unfolds: arrests are made, a re-examination is announced, an inquiry committee is constituted, Parliament erupts in outrage. Then the storm passes, the controversy fades — and the next examination cycle brings the same violation through the same vulnerabilities. The reason is plain: the problem does not lie with the culprit alone; the problem lies at the root of the system. Nobody goes back to fix the weaknesses that made the leak possible in the first place.I write this not as a political commentator, but as a teacher with direct experience inside examination administration. I am a professor at a Nanded university, and have been actively associated with the university’s examination department. I have personally managed the conduct of degree and postgraduate examinations for approximately two to three lakh students across more than a hundred examination centres. I have witnessed the confusion when a wrong question paper reaches a centre, the anxiety among students when printing is delayed, and the errors that distort results when answer-sheet scanning goes wrong. I know this system’s weak points from the inside — and I have thought carefully about how to strengthen them.Every paper leak has one common thread: somewhere, before the examination began, a readable question paper reached human hands. Eliminate that access — and you eliminate the leak. With the technology India already possesses, this is well within reach.The answer is not harsher punishment, even though those found guilty must face the severest penalties the law permits, and a transparent and impenetrable examination system must be built as a permanent deterrent. Nor is a wholesale shift to online examinations any kind of solution. The risks of server hacking and remote access, combined with the disadvantage it places on students from rural areas who are less familiar with digital technology, simply create a new set of problems in place of the old. The answer is a fundamental redesign of the entire examination process, phase by phase. I have named this framework the Secure National Examination Conduction System or SNECS.SNECS is built on one clear principle: no question paper, in physical or digital form, shall be accessible to anyone, anywhere in the country, until it is printed automatically inside a locked and strictly monitored examination control room, between 30 and 60 minutes before the examination begins. Until that precise moment, the question paper does not exist in any readable form, anywhere.How is this achieved? No human expert sets the question paper in the traditional sense. Instead, Artificial Intelligence automatically selects the question paper from a secure digital repository of five to ten lakh validated questions. Subject experts contribute questions to this repository, and translators render them into regional Indian languages; but neither group will ever know which questions were finally selected, because that selection is made entirely by the system, not by a person. The final paper is generated in four differently sequenced sets and locked in high-grade encryption. To unlock it, five independent institutions must act together – the NTA, the Ministry of Education, the CBI, a Supreme Court-appointed observer, and the state government – and any three of the five must authorise simultaneously before the paper can be decrypted and printed. Even if corrupt elements were to infiltrate the NTA itself, they could not open the paper alone.Each examination centre has a specially secured printer, comparable in its level of protection to a bank ATM. This printer, upon receiving the authorisation signal, prints the question paper automatically. It cannot display the paper on any screen; it can only print. Candidates are admitted through a three-level biometric verification process: facial recognition at the entrance gate using a biometric machine, a physical check for prohibited items, and fingerprint authentication at the examination seat. No candidate can impersonate another and pass all three checks.It is worth being explicit about what biometric verification does – and does not – do in this context. Unlike surveillance systems that attempt to identify unknown individuals from imperfect, often degraded, images with the attendant risks of misidentification and wrongful exclusion, SNECS uses biometrics only to confirm that the person presenting themselves at the gate is the same individual who voluntarily enrolled weeks earlier. A mismatch is not a gate that closes. It is a flag that triggers human verification: a Centre Technical Supervisor reviews the case, a secondary iris scan is offered as an alternative, and a invigilation officer is physically present on site. No student is turned away on the basis of algorithmic uncertainty alone. For candidates with worn fingerprints, physical disabilities, or failed biometric scans, alternative authentication channels should be made available. The system’s purpose is not surveillance, it is to end organised impersonation rackets, which harm, above all, the sincere student from an ordinary background whose seat is taken by a paid proxy.Admit cards are issued three days before the examination, so that neither candidates nor supervisory staff know their assigned centre well in advance. Supervisors and support staff are assigned to their centres by computer-generated random selection, no more than 48 hours before examination day, and no supervisor is posted to a centre in their own city. This leaves no time – and no opportunity – to build the local relationships that centre-level leaks depend upon. Flying squads make unannounced visits to centres throughout the examination, with full authority to cancel proceedings and preserve evidence on the spot.Immediately after the examination concludes, OMR answer sheets at every centre are scanned by portable high-speed scanners, and the data transmitted directly to NTA’s secure servers. Provisional results are declared within 24 hours. This speed is not merely a matter of efficiency but an anti-corruption mechanism. When results are declared this quickly, there is simply no window left for anyone to intercept, alter, and return answer sheets before the outcome is already public knowledge.SNECS requires no foreign or untested technology. Aadhaar biometric authentication already serves over 133 crore citizens. DigiLocker, the NIC Cloud, income-tax e-filing, Aadhaar-linked public distribution, digital payments, and India’s wider e-governance infrastructure — these are proven, fully operational systems. SNECS simply brings the relevant components of this existing ecosystem together and applies them to the conduct of examinations. The estimated first-year cost of implementation is approximately Rs 300 to 420 crore. By comparison, the economic harm caused to twenty-two lakh students by the NEET 2026 re-examination alone — in travel, accommodation, lost preparation time, and deferred admissions — is conservatively estimated at Rs 660 to 1,760 crore. SNECS recovers its entire implementation cost within a single examination cycle.The Parliamentary Standing Committee on Education has summoned the NTA chairperson. The Supreme Court has intervened repeatedly. The CBI is conducting raids across multiple states. All of this is necessary but none of it is sufficient. These measures address the symptom, the leak, rather than the underlying condition that makes leaks inevitable: a system in which too many human beings must handle the question paper between the moment it is created and the moment a student sees it. Wherever human hands touch the question paper, a point of vulnerability exists. Reducing that contact to the absolute minimum is the only lasting solution.India has landed a spacecraft with precision on the south pole of the moon. India has built a digital payments infrastructure. Can India not build an examination system free from paper leaks for 25 lakh students? This is not a question of technological capacity but of priorities.The students of this country have earned a better system. Dr. Santosh Ramrao Butle is a Professor in the School of Pharmacy at Swami Ramanand Teerth Marathwada University (SRTMUN), Nanded, Maharashtra. He has been actively associated with the university’s examination department and has direct operational experience managing hybrid-mode examinations for approximately two to three lakh students across more than a hundred centres annually. He has developed the Secure National Examination Conduction System (SNECS) and submitted it as a formal policy proposal to the National Testing Agency, the Ministry of Education, Government of India, and the Parliamentary Standing Committee on Education.