Everyone's talking about the hack. The five elementary vulnerabilities in the re-evaluation portal. The security researcher who walked straight through the door. That part of the story is genuinely alarming — but it's also a distraction from something worse.
The hack tells you that CBSE has poor cybersecurity hygiene. The procurement story tells you something about how decisions get made inside the institution, whose warnings get ignored, and who ends up paying the price when it all goes wrong.
The answer to that last question is 1.8 million Class 12 students.
Nobody Wanted This Contract
Before a single answer book was scanned, the market sent CBSE a clear message: this is a bad idea.
CBSE floated tenders for the On-Screen Marking system — the platform that would digitise the evaluation of 1.8 million students' answer books — and found no qualified bidder. Then it tried again. Same result. It took a third attempt before a company called Coempt EduTeck qualified and won the contract. CBSE's own public statement referred to "the qualified bidder" — singular, carefully worded — not the best bidder, not the top-scoring applicant. The one who qualified. After multiple rounds that produced no winner.
When a government tender fails to attract qualified bidders repeatedly, that's the market telling you something. Either the specifications are poorly designed, the budget is too low, or the technical requirements are unrealistic. The correct response is to pause, reassess, and fix the problem. CBSE's response was to keep running the tender until someone — anyone — cleared the bar.
What changed between the failed attempts and the successful one? Did the eligibility criteria shift? Did the financial thresholds drop? Did the technical requirements get simplified? CBSE has not answered these questions. It has not even been publicly named Coempt EduTeck as the vendor — the company appears on Coempt's own website, but CBSE has never officially confirmed it.
CBSE's Own Governing Body Said: Don't Do This Yet
This is the part of the story that has received almost no attention, and it's the most damning detail of all.
In June 2025, CBSE's governing body — its highest internal oversight body — recommended that OSM:
May be implemented in all subjects only after completion of pilot projects in some subjects across various regional offices of the board.
CBSE has 22 regional offices. The governing body was saying: Test this properly before you roll it out to the whole country.
Zero pilots were conducted. None.
Instead, CBSE ran a two-day dry run in January at five Delhi schools with 100 teachers. The teachers who took part told journalists they had advised CBSE against proceeding — they cited inadequate features, insufficient training, and a need for more time to adapt. CBSE proceeded anyway.
The formal rollout was announced on 9 February. Class 12 examinations began on 17 February. The training portal for evaluators opened on 15 February — four days before results were due to be marked.
Think about that sequence for a moment. The governing body said run regional pilots. CBSE ran a two-day exercise at five schools in one city. The teachers in that exercise said stop. CBSE announced the rollout six days before exams started. Evaluators were being trained on the software while students were taking the papers that the software would grade.
The Failure That Was Predicted, Then Delivered
Once the system went live, it performed exactly as warned.
Officials acknowledged login failures, server overloads, and scanning deficiencies. Of 9,866,622 answer books evaluated this year, 68,018 had to be rescanned due to poor image quality. Another 13,583 were checked manually after repeated scanning failed entirely. That's nearly 82,000 answer books — roughly the population of a mid-sized town — that the system could not handle correctly.
One mathematics teacher described how fatigue from hours of screen-based work affected step-marking accuracy. "Some evaluators were still figuring out the software while checking live answer books," the teacher said. A school principal reported that teachers were under pressure to meet daily targets so results could be declared on time. Speed mattered more than careful reading.
This is what happens when you deploy an untested system at a national scale under time pressure, against the explicit advice of your own governing body, your own dry-run participants, and the signal sent by a market that rejected your contract twice.
The Public Noticed What CBSE Wouldn't Admit
By 26 May, CBSE had received 404,319 applications seeking scanned copies of 1,131,961 Class 12 answer books. That represents a jump of more than 208 per cent in applications and 301 per cent in answer-book requests compared to the previous year.
CBSE attributed part of the surge to a fee reduction — from ₹700 to ₹100 per subject. A price cut would explain some increase. It does not explain a tripling in the number of answer books being requested. Students and parents are not taking advantage of a discount. They are expressing systemic distrust in the results they received.
The overall Class 12 pass percentage fell 3.19 percentage points to 85.20 per cent — the lowest since 2019. Three explanations exist: students performed worse, OSM marked more harshly, or OSM made errors. Given that evaluators were learning the software in real time, under speed pressure, on a system the governing body had flagged as unready, the third explanation cannot be dismissed. CBSE has not dismissed it either. It has simply said the system will continue next year.
This Wasn't Bad Luck
Lay out the sequence end to end, and the picture becomes coherent, even if uncomfortable.
Three failed tenders. A vendor with a history of problems in the same category of work. A governing body recommendation was ignored. A two-day dry run at five schools, substituting for nationwide regional pilots. A system was announced six days before the exams. A training portal opened four days before marking began. 82,000 answer books the system couldn't process. A 3.19 percentage point drop in the pass rate. A 301 per cent surge in re-evaluation requests. And CBSE has still not officially named the company that built the platform.
This is not a run of unfortunate coincidences. This is an institution that received warnings at every level — from the market, from its own governing body, from the teachers in its own dry run — and chose to proceed. The question is why. Was it political pressure to demonstrate digitisation? A contractual deadline that couldn't slip? An institutional inability to admit that the project wasn't ready?
CBSE hasn't explained. It has acknowledged "early technical problems," opened a re-evaluation portal, commissioned an IIT audit, and waited for the news cycle to move on.
It has not answered the most basic question the procurement record raises: if nobody qualified in the first two rounds, what changed in the third? What exactly made Coempt EduTeck "the qualified bidder" after two rounds that produced no one?
1.8 million students would like to know.
