In May, an eighteen-year-old Class 12 student made an unsettling discovery. While accessing the digital verification window under the CBSE's newly mandated On-Screen Marking (OSM) system, he found that his physics paper bore handwriting that was not his. Frustrated by silence from official helplines, he took to X. His post went viral, and became the defining moment of a national controversy that has since escalated from administrative embarrassment to judicial challenge.
It is a controversy that cuts to the heart of an ineluctable question facing India's largest examination board: when technology is deployed at a scale affecting millions of young lives, who bears the cost of its failures? While the education ministry sacked the Board’s chairman and secretary, there was little compensation for the mental agony of the students.
What’s OSM
OSM is a digital evaluation framework in which physical answer scripts are scanned at designated hubs, anonymised to conceal student identities, uploaded to a secure server, and distributed electronically to evaluators. Teachers mark the scanned scripts on computer screens and submit their scores digitally. The system eliminates paper transport, shortens the timeline, and, in theory, eradicates totalling errors and handwriting-based bias. CBSE Examination Controller Sanyam Bhardwaj announced in February last that the board had decided to introduce OSM for Class 12 answer books beginning with the 2026 examination, in its continuous effort to enhance efficiency and transparency, while Class 10 evaluation would continue in physical mode. The promise, as articulated by the board, was a faster, fairer, more transparent system of assessment. However, in reality, it was anything but.
The scale of the exercise was unprecedented in Indian examination history. According to officials, over 98 lakh answer books, amounting to nearly 1.96 crore scanned pages, were digitally evaluated. Of these, 68,018 answer books had to be rescanned due to poor image quality, while 13,583 answer books were manually checked after repeated scanning attempts failed to produce readable copies.
The procurement trail: Hubris before the fall
The origins of this conflagration lie not in May 2026 but considerably earlier. CBSE floated three separate tenders for the OSM system before finally selecting a vendor. The first tender received no bids. The second failed to produce a technically eligible bidder. The board eventually modified several technical requirements in a third tender issued in August 2025, just six months before its implementation. The tender-bender was flagged by a student whistleblower on his blog, which went viral. He subsequently appeared before a parliamentary committee to share his findings. The contract for the digital platform was awarded just 74 days before the board exam began on February 17.
The Board's tryst with digital evaluation is not recent. The CBSE first conceptualised OSM in 2014, before dropping it as contemporary scanners required cutting booklet spines to feed loose sheets into rollers. The plan was revived a decade later with book-scanning technology. The Visvesvaraya Technological University in Karnataka had already traversed this terrain, deploying OSM for engineering semester examinations as early as 2011–12 to curb endemic tampering. VTU's experience offered a mixed preview: result-processing time shrank, but the system drew sustained criticism for poor scan quality.
The warnings that were overridden
The documented trail of warnings that preceded the rollout suggests institutional failure. In January 2026, CBSE conducted a pilot of the OSM system in Delhi involving principals, examiners, evaluators and subject matter experts from government, private, Kendriya Vidyalaya and Navodaya schools. Participants are said to have identified multiple technical and operational concerns and recommended that the platform undergo at least one more year of trial and refinement before national implementation. Two reports were subsequently submitted to the Board. Many of the complaints raised by students after the declaration of results, including concerns over scan quality, discrepancies in evaluation and portal malfunctions, echoed issues highlighted during the pilot exercise. Rather than deferring the rollout, the CBSE proceeded with additional webinars and practice sessions for evaluators.
The Delhi Government School Teachers' Association, too, urged the CBSE to implement OSM from the next session, stating that implementing a fully digital evaluation system without adequate preparation and structured training presents significant practical challenges. The advice fell on deaf ears.
Three structural failures
The problems that surfaced during the evaluation cycle may have reflected deeper structural challenges rather than isolated technical glitches. One issue concerned scan fidelity. CBSE acknowledged that tens of thousands of answer books had to be rescanned because of image-quality problems, while thousands more required manual evaluation. Student complaints regarding illegible scans and missing responses suggested that preserving the clarity of diagrams and handwritten answers at such scale posed considerable difficulties. A second challenge lay in the uneven technological infrastructure across evaluation centres. CBSE prescribed minimum requirements, including specified operating systems, internet connectivity, and uninterrupted power supply. Yet the quality of hardware and network conditions available to schools varied considerably, raising questions about the assumption that a uniform digital platform could function identically across diverse institutional settings.
A third challenge stemmed from changes in the evaluation process itself. Under the traditional system, evaluators worked together at physical hubs and could consult Head Examiners in real time when confronted with ambiguous or unconventional answers. The transition to digital assessment altered these patterns of interaction, replacing a more collaborative environment with a standardised electronic workflow. Whether this shift affected the quality and consistency of evaluation merits closer examination.
The global mirror
The international experience on OSM offers a sobering compendium of lessons that CBSE appears not to have absorbed. In the United Kingdom, examination boards including AQA, OCR, and Pearson Edexcel have operated online marking for GCSE and A-Level evaluations for over a decade. The International Baccalaureate, operating across more than 150 countries, uses centralised digital scanning networks monitored by senior moderators via live dashboards. These systems succeed partly through a practice called item-level marking: scanned scripts are sliced into isolated questions, each assessed by a different evaluator, creating inherent standardisation and preventing any single examiner's fatigue or bias from contaminating a whole paper. A second safeguard is the seed script, a pre-marked paper injected silently into an evaluator's digital queue. If the teacher scores the seed script inaccurately, the system locks them out pending recalibration. Both mechanisms demand sophisticated software architecture, extended pilots, and iterative refinement. CBSE deployed neither at national scale.
Scotland's 2000 debacle under the Scottish Qualifications Authority is the cautionary archetype. In an effort to automate the processing and tracking of new curricular assessments, the SQA's systems suffered a conflagration of catastrophic data failures: thousands of incorrect certificates were issued, and approximately 5% of schools received no results at all on results day. The crisis became a textbook case in the ramifications of deploying complex digital systems at national scale without adequate stress-testing. CBSE's dry-run data pointed to precisely those failure modes, yet the Board chose to go ahead regardless.
Compounding the technical failures is a policy decision that became deeply contentious: the CBSE abolished post-result verification of marks for Class 12 as part of the OSM rollout. It said students would no longer be able to apply for post-result verification of marks for Class 12, confident that improved accuracy would make it unnecessary. The OSM bugs subsequently forced the Board to open the window for verification of scores.
On June 2, 2026, a Public Interest Litigation was filed in the Delhi High Court, to be heard by a vacation bench on June 8, demanding an independent inquiry into the large-scale irregularities.
The way forward: Rectitude over speed
The government has since constituted a one-member committee headed by S Radha Chauhan to investigate the award and procurement of services related to the OSM platform, and submit a report within a month. For OSM to earn the trust it currently lacks, the CBSE must review its position that speed of result declaration is the primary metric of success.
Resolution-specific scanning standards must be mandated, with a minimum 300 DPI grayscale scan for all subjects containing visual and mathematical content, and with automated edge-detection software to flag and reject blurred or low-contrast pages before they reach any evaluator's queue. This is not a chimera; it is standard practice in forensic document scanning and should have been built into the OSM specification from the outset.
Dual-verification checks must be embedded in the submission workflow, requiring each evaluator to confirm a barcode match against an independent physical tally before any script is finalised, making student-profile mismatch errors mathematically impossible rather than merely improbable. Item-level marking and seed scripts, as deployed in the UK for over a decade, must be integrated into the system architecture rather than treated as otiose add-ons for a future phase.
A hybrid revaluation mechanism in which students disputing a scan's fidelity can request physical re-examination of the original script preserved at Regional Offices is not incompatible with digital evaluation. It is its essential safeguard.
The OSM rollout will be remembered as an important moment in Indian education policy because implementation and integration of large-scale digitisation demand multiple layers of checks and balances. Technology deployed in haste, at this scale, on students this young, with stakes this high, is not modernisation. It is recklessness dressed as reform.
Tender process
CBSE issued three tenders before selecting a vendor for the OSM project The first tender received no bids The second tender failed to produce a technically qualified bidder Technical requirements were modified in a third tender issued in August 2025 The nationwide rollout followed within approximately six months.
Warnings before implementation
CBSE conducted a pilot exercise in Delhi in January 2026 The pilot involved principals, examiners, evaluators, and subject experts from government and private schools, Kendriya Vidyalayas, and Navodaya schools.
Participants identified numerous technical and operational concerns. Two reports were submitted to the Board after the exercise Participants recommended additional testing and refinement before nationwide implementation Teachers' associations also urged CBSE to postpone the rollout.
Structural challenges
Tens of thousands of answer books had to be rescanned because of image-quality problems Preserving the clarity of handwritten responses and diagrams at such scale proved difficult.
Evaluation centres operated under diverse hardware and connectivity conditions despite uniform technical requirements. Digital evaluation altered traditional patterns of interaction between evaluators and Head Examiners.
Questions arose regarding whether the new workflow affected consistency and quality of assessment.
