Five Standards India's On-Screen Marking Framework Must Mandate After CBSE 2026

The Gap the Supreme Court Identified

On July 15, 2026, India's Supreme Court heard a PIL that explicitly asked for "regulations governing on-screen marking." The bench, led by Chief Justice Surya Kant, noted "creeping problems" in the existing digital evaluation system and directed the Solicitor General to submit a status report by July 24.

That phrase — "regulations governing on-screen marking" — is the operative one.

Currently, no formal regulatory standard governs how examination bodies in India must design, procure, or operate OSM systems. CBSE issued its own internal circular in February 2026 describing its system. State boards devise independent approaches. Universities follow their own conventions. There is no shared technical standard, no minimum scanning quality threshold, no mandatory audit trail specification, and no common student rights framework.

India's examination evaluation infrastructure processes over 4 crore answer scripts annually. The absence of a regulatory floor for the technology handling those scripts is no longer defensible. The Radha Chauhan one-member commission, constituted through the Cabinet Secretariat to review CBSE's vendor procurement, will almost certainly surface this gap in its findings.

What should good OSM regulation look like? Based on the CBSE 2026 experience and international practice from Cambridge Assessment, Pearson Edexcel, and the UK's Ofqual, five standards emerge as the essential foundation.

Standard 1: Minimum Scanning Quality Specification

The problem: CBSE scanned 98 lakh answer sheets and discovered 68,000 with quality issues. Some were not identified until the evaluation process was already underway. A further 13,000 scripts never reached legible quality even after rescanning and required manual assessment.

What the standard should require: All scanning operations must produce images at a minimum of 300 DPI for legible handwritten text. A machine-readable legibility check must be conducted on every batch of scanned sheets before that batch enters the evaluation queue. No script may be assigned to an evaluator until it passes the quality threshold. Scripts that fail twice must be flagged for manual assessment and separately tracked.

Examination bodies must publish their scanning quality benchmarks — minimum DPI, batch failure rate, rescan rate — as part of their result declaration report each year. Transparency on these numbers should be non-negotiable.

International precedent: Cambridge Assessment specifies minimum resolution standards in its OSM vendor contracts and conducts pre-evaluation quality audits for every subject cohort. The UK's Ofqual includes scanning quality as a monitored parameter in its examination system oversight.

Standard 2: Pre-Flight Identity Verification

The problem: Twenty CBSE answer sheets were found to contain the wrong student's content. Students discovered these mix-ups themselves by accessing the portal — not through any internal quality process operated by CBSE. This is a structural failure: the institution that administered the examination had no mechanism to detect its own identity errors before students pointed them out.

What the standard should require: Before any scanned script enters the OSM evaluation queue, an automated verification step must confirm that the barcode or QR code on the script matches the student registration record in the examination database. Any script where the identity marker is damaged, absent, or unreadable must be held out of the digital evaluation queue and escalated for manual verification. The outcome of every identity check must be logged.

A secondary verification — comparing the script's handwritten roll number (where present) against the barcode value — provides a redundant check. Scripts where the two do not match must never proceed to evaluation without manual review.

Standard 3: Evaluator Certification Before Platform Access

The problem: CBSE's first-year roll-out exposed a significant training gap. The OSM interface presents one question at a time to an evaluator — a fundamentally different cognitive and procedural experience from reviewing a physical answer book. Evaluators unfamiliar with the interface showed inconsistent marking behaviour. The platform cannot compensate for evaluators who have not been trained on it.

What the standard should require: No evaluator may mark on an OSM platform without completing a minimum six-hour certified training programme. This must include a mock-evaluation session using historical scripts with known correct answers. The system must track each evaluator's average marking time per question, flag responses that deviate significantly from the evaluator's own mean, and generate evaluator performance reports that the examination body can audit.

Evaluators who fail the mock evaluation must not be assigned to live scripts. Examination bodies must retain records of evaluator certifications and make them available to accreditation bodies on request.

Standard 4: Immutable Audit Trail for Every Marking Action

The problem: A persistent concern from students and courts in 2026 has been whether specific questions were evaluated or skipped in the digital system. This question should have an unambiguous, retrievable answer from an audit log. In the CBSE 2026 controversy, the absence of easily accessible audit documentation deepened public distrust.

What the standard should require: Every marking event — evaluator opens a question image, submits a mark, navigates to the next question — must be time-stamped and written to a tamper-proof log. The log must use a cryptographic hash chain or equivalent mechanism to prevent post-hoc alteration. It must be stored separately from the evaluation interface and accessible to the examination authority independently of the OSM vendor.

Students with active re-evaluation applications must have a right to receive a printed extract of the relevant section of the audit log showing whether their question was opened and marked. Audit logs must be retained for a minimum of seven years.

Standard 5: Student Visibility and Dispute Resolution Timelines

The problem: In the CBSE OSM cycle, students could view their scanned answer sheets on the portal. This transparency was largely accidental — not mandated by policy — but it was also what allowed the mix-up errors to surface. The dispute resolution process that followed operated without clear timelines, under judicial pressure, and inconsistently across subjects.

What the standard should require: Every student must be able to view their authenticated scanned answer sheet within 48 hours of result declaration, at no cost. Any dispute relating to an incorrectly assigned script must be resolved within 10 working days of the dispute being filed.

Within 30 days of result declaration, the examination body must publish aggregate data on: number of scripts rescanned, number of identity disputes, number of scripts reviewed under re-evaluation, and percentage of marks changed on re-evaluation by subject. This data must be publicly accessible and not require an RTI application to obtain.

A Compliance Table for Examination Bodies

The following table maps each standard to what university examination bodies should incorporate into platform procurement and vendor contracts, whether or not formal regulation has arrived.

The Regulatory Window Is Now

The Radha Chauhan commission's findings and the Supreme Court's July 24 order will together frame whatever formal OSM regulations emerge over the next 12 to 24 months. The Ministry of Education, CBSE, and state examination boards will be the direct addressees of those regulations.

But universities and autonomous colleges do not need to wait. The five standards above reflect what careful examination bodies already do — or what they would do if asked to justify their OSM system in a court of law. Building these standards in now means that when formal regulation arrives, it will confirm what is already in place rather than require a disruptive retrofit.

India's examination technology sector is at a turning point. The Supreme Court's intervention is not a verdict against digital evaluation. It is a demand that digital evaluation meet the bar that public trust requires. Meeting that bar — through scanning quality, identity verification, evaluator training, audit trails, and student transparency — is the work of institutions, not courts.

Related Reading