Industry2026-09-17·8 min read

90 Exam Scandals in 20 Years: Why Tamper-Proof Evaluation Is Now a Risk Management Imperative

India has recorded more than 90 examination irregularities and paper leaks in the past two decades. The statistical case for investing in tamper-proof digital evaluation infrastructure has never been stronger.

A Number That Should Change How Universities Think About Risk

Ninety. That is the approximate count of documented examination scandals, paper leaks, and evaluation irregularities at major national and state examination bodies across India over the past two decades, according to reporting from The Tribune India. Of those, at least fifteen involved examinations conducted by central agencies — NEET, JEE, UGC NET, and their counterparts. The rest happened at state boards, university examination cells, and recruitment commissions across the country.

2026 delivered two of the most consequential incidents in that record. On May 3, the National Eligibility cum Entrance Test — NEET-UG — was administered to 22.79 lakh candidates. Eleven days later, on May 14, the government announced a full cancellation. A "guess paper" containing approximately 120 of the 410 questions had circulated weeks before the examination. The re-test that followed was one of the largest re-examinations in Indian history. A 23-year-old medical aspirant died by suicide in the days after the cancellation.

In May and June, CBSE's first full-scale On-Screen Marking cycle for Class 12 generated over 70,433 formal student grievances, triggered the removal of the Board's Chairman and Secretary, and prompted a parliamentary inquiry.

The question for university administrators is not whether these events are relevant to them. The question is whether their institutions have built the infrastructure to survive when a comparable incident hits.

The Base Rate Problem

Risk in institutional examination management is not random. It has a base rate — a background probability that, given enough cycles, an incident will occur. The national data suggests that base rate is not trivial.

India conducts examination cycles at a frequency that makes individual incident probability nearly irrelevant. A typical affiliated university running two full evaluation cycles per year over a ten-year operating period completes twenty evaluation windows. A state board running four windows annually completes forty over the same period.

At the documented national incident rate — ninety events spread across thousands of institutional examination cycles over twenty years — the individual annual probability may appear low. But the cumulative probability over twenty cycles is not. And when an incident does occur, whether it involves a lost answer bundle, a marks inflation allegation, a tampering complaint, or a data integrity failure in a digital system, the institution faces an immediate question it cannot defer: what does your audit trail show?

An institution with comprehensive digital evaluation records has an answer. An institution operating on paper-based or partially digital systems frequently does not.

The Evaluation Chain: The Unsecured Half of Exam Security

Public discourse about examination security focuses almost exclusively on the question paper — the leak before the exam. The Anti-Paper Leak Amendment Act, Fast-Track Courts, and multi-agency investigations are all calibrated to answer the question of how question papers escape before examination day.

They are calibrated for one half of the problem.

After students put down their pens, answer scripts enter a separate physical and administrative chain that receives almost no equivalent scrutiny. In a paper-based system, that chain includes:

Physical transport of sealed answer bundles from examination halls to collection points

Storage pending the evaluation window

Distribution of individual answer books to evaluators at evaluation camps or home-marking arrangements

Return of evaluated scripts to the marks tabulation centre

Manual entry of marks into a register or digital system

Verification of totalling before result declaration

At each node, there is exposure. Bundles can be lost or damaged in transit. Individual answer books can be removed and replaced. Marks can be altered between evaluation and tabulation. Re-evaluation processes can be manipulated. When a student challenges a declared result, the institution must reconstruct what happened at every one of those nodes — an exercise that is frequently impossible without digital records.

A 2026 research compilation on examination security failures found that a significant proportion of university-level evaluation disputes stem not from question paper leaks but from evaluation chain failures: missing answer books, marks transcription errors, unexplained differences between original evaluation and moderation outcomes, and in some documented cases, tampering allegations at the re-evaluation stage.

What Tamper-Proof Digital Evaluation Actually Delivers

The Nilekani Task Force, which submitted its final report on September 13, 2026, recommended minimum technical standards for digital evaluation platforms that address the evaluation chain specifically. Those standards reflect what secure evaluation infrastructure must provide:

Hash-verified scanned images. Every scanned answer book page receives a cryptographic hash at the time of scanning. Any modification to the image file — even a single pixel — invalidates the hash and is immediately detectable. Physical answer books can be altered; hash-verified digital images cannot.

Evaluator session logs. A complete audit trail records every mark awarded, every question viewed, the duration of time spent on each page, and the identity of the evaluator who made each entry. This is not just useful for dispute resolution — it is a forensic record that makes mark tampering after evaluation technically implausible.

Double valuation with programmatic moderation. When two independent evaluators assess the same answer script without seeing each other's marks, and an automated system triggers moderation only when their scores diverge beyond a configured threshold, the system eliminates both conscious bias and unintentional error simultaneously. This process, which is expensive and logistically complex in paper-based evaluation, runs as a workflow configuration in a digital platform.

Immutable re-evaluation records. When a student applies for re-evaluation, the digital system presents the original evaluator's mark, the second evaluator's mark, and the moderator's decision — all timestamped and identity-linked. The re-evaluation outcome is recorded against the original record, not in its place. There is no mechanism through which a re-evaluation can erase the original evidence.

Marks tabulation from evaluation data, not manual entry. Because marks flow directly from the evaluator's digital interface to the marks database, the manual transcription error that is one of the most common sources of incorrect results in paper-based systems is eliminated by design.

What the 90-Incident Record Means for Institutional Risk

An institution that has not experienced an examination integrity challenge is not necessarily operating safely. It may simply not have had its incident yet — or may not have detected the ones that did occur.

The distribution of examination scandals across India suggests several structural factors that university examination cells share with documented incident sites:

Seasonal evaluation concentration (all exams evaluated in a compressed window with large numbers of evaluators hired on short tenure)

Physical answer book handling with minimal chain-of-custody documentation

Manual marks entry as the final step before result compilation

Re-evaluation processes that allow substitution rather than supplementation of original records

These are not failures of individual institutions. They are characteristics of paper-based examination management at scale. And they are characteristics that digital evaluation infrastructure eliminates.

The Institutional Protection Argument

When an examination dispute reaches a court, a tribunal, or an RTI proceeding, the institution that can produce a complete, timestamped, identity-linked record of every step in the evaluation chain is in an entirely different legal and reputational position from one that cannot.

The Supreme Court has repeatedly engaged with CBSE's OSM controversy during 2026, specifically in the context of what records the Board could produce when student grievances were filed. In one set of proceedings, the court directed CBSE to make per-question evaluation records available to students who had filed revaluation applications — a direction that was only actionable because the digital system had captured that data. An institution without equivalent records would face a materially different inquiry.

Courts do not distinguish between an institution that actively suppressed records and one that simply did not generate them. The outcome for the institution — inability to refute allegations, adverse inferences, and reputational damage — is similar in both cases.

Where to Start

The CBSE experience and the Nilekani Task Force recommendations together define what secure evaluation infrastructure looks like. Universities that have not yet adopted digital evaluation do not need to replicate CBSE's scale on day one. They need to address the evaluation chain systematically, beginning with the highest-exposure nodes:

Scanning with hash verification — eliminate physical answer books as the single source of truth

Identity-linked evaluator access — every mark is attributed to a specific, authenticated individual

Double valuation as a workflow default — not an exception for contested results, but the standard process

Direct marks flow to the result database — no manual transcription step

Per-question audit records accessible to students — the UGC's 2026 answer script access guidelines require this

Ninety incidents in twenty years is not a trend that will reverse without structural change. The institutions that build tamper-proof evaluation infrastructure now are not just making a technology investment — they are making a risk management decision whose value becomes clear the day an incident occurs and they can demonstrate, record by record, exactly what happened.