9
The directions the Supreme Court issued on 15 January 2026, in exercise of its powers under Article 142, on student deaths in higher education. Four of the nine concern data and reporting; the rest concern medical access, staffing, scholarships and compliance with the regulations that bind each institution.
Student Deaths in Higher Education: What Is Required, and What Can Actually Be Checked
Student deaths in India's higher education institutions raise questions that the public record does not yet answer with a single verifiable count.
Before anything else, the limits of this page. Three of 23 IITs audited: 9 cases at Delhi, 11 at Madras, 9 at Kanpur. One of the Delhi nine is provisional — a postgraduate student who died in August 2026, with no institute statement and no police classification at the time of the audit, which our own Delhi report says must be re-checked before publication. That re-check is not documented. Our cross-institute table counts 9 for Delhi; the Delhi report's own certification block counts 8, excluding that case. We publish 9 and name the difference rather than resolving it quietly. Nothing here is a confirmed suicide finding, and nothing here should be added into a national figure.
Several figures circulate. Ministerial answers to Parliament give all-IIT totals for various periods. RTI replies give institute-level counts on academic years. An independent compilation gives a much larger cumulative figure, built partly from RTI responses covering fewer than half the IITs. These are not competing counts of one thing — they cover different years, different institutions, and different definitions of who counts as a student. Added together they produce a number that means nothing, and no member of the public can reconcile them.
In January 2026 the Supreme Court moved this forward significantly. Exercising its powers under Article 142, it required suicide data for ages 15–29 to be centrally maintained; required the National Crime Records Bureau to distinguish school students from higher-education students; required every institution to report any student suicide or unnatural death to the police as soon as it learns of it, wherever the death occurred and whatever mode the student was studying in; and required an annual report of those deaths to the University Grants Commission and the professional regulators — or, for Central Universities and Institutes of National Importance such as the IITs, to the Department of Higher Education. That machinery is now binding.
But nothing in those directions requires any of it to reach the public. The reports go to a regulator or a ministry, not to citizens. There is no shared definition of "student", no shared vocabulary for classifying a death, no case identifier that lets one record be followed from an institute to a ministry to a national total, no obligation for those totals to agree, no history when a figure changes, and no independent audit of any of it.
We know what that gap costs, because we measured it. We audited three IITs — Delhi, Madras and Kanpur — building each case list independently and then testing it against every official claim we could find. Of the six institutional, RTI and Commission-level claims we examined at Madras and Kanpur, none held up as originally framed. One cited ruling turned out to reach the opposite conclusion from what was claimed of it. And across all three institutes, not one primary document was retrieved in full for any individual death.
That is not an allegation of concealment. We found no evidence of that and we do not claim it. It is a description of a record that cannot be checked — and a figure that cannot be checked cannot support accountability in any direction.
So we are not asking for this data to be collected or reported. The Court has already required both. We are asking for the resulting record to be published in a form that can be verified, and we have set out exactly what that standard should be.
0
of those nine directions require the resulting record to be published where the public can check it. The word does not appear in them.
3 of 23
IITs have been audited in depth by Aarunya so far. The remaining 20 have not been examined — not audited does not mean zero.
3.5%
of India's 60,383 higher education institutions responded to the Supreme Court task force's own survey. The figure is the Court's, recorded in the January 2026 judgment.
What the Supreme Court already requires
On 15 January 2026, in Amit Kumar & Ors. v. Union of India & Ors. (2026 INSC 62), the Supreme Court issued nine directions "in exercise of our plenary powers under Article 142 of the Constitution of India." Four concern data and reporting. In the Court's own words:
(i) "The Sample Registration System data on Suicides, especially those falling within the age group of 15-29 years, must be centrally maintained for better and more accurate estimates of Deaths by Suicide of students in HEIs."
(ii) "The NCRB, in its annual report, must distinguish between school-going students and students of higher education in its categorisation of 'student suicides'…"
(iii) "All HEIs must report any incident of suicide or unnatural death of a student, regardless of the location of its occurrence (i.e. on campus, hostels, PG accommodations, or otherwise outside the institutional premises), to the police authorities no sooner they come to know about the incident. This should cover all students - irrespective of whether they are studying in the classroom, distance or online mode of learning."
(iv) "…an annual report of student suicides or unnatural deaths must also be submitted to the UGC & all other relevant regulatory bodies for professional courses (e.g. AICTE, NMC, DCI, BCI etc.). In case of Central Universities and Institutes of National Importance… or any HEI that does not fall within the above-mentioned framework, it must be reported to the Department of Higher Education, Ministry of Education, Government of India."
The other five concern institutional conditions: round-the-clock qualified medical help on campus or within one kilometre for every residential institution; all vacant faculty positions, both teaching and non-teaching, filled within four months with priority to reserved posts; Vice-Chancellor, Registrar and other key posts filled within four months, with annual reporting to central and state governments on how many reserved posts are vacant, filled, and why; scholarship backlogs cleared within four months, with no student barred from an examination, a hostel or a class because of a disbursement delay; and strict compliance with the UGC regulations on ragging (2009), equity (2012), sexual harassment (2016) and student grievance redressal (2023).
Two things follow, and they matter for what comes next. The IITs are Institutes of National Importance, so their annual reports go to the Department of Higher Education. And nothing in these nine directions requires any of this to be published — we searched the full judgment; the word does not appear in that sense anywhere in them.
Everything in the next section is Aarunya's proposal. The Court has ordered none of it.
A second judgment the institutions also answer to
The nine directions of January 2026 are not the only set an institution has to satisfy. Six months earlier, on 25 July 2025, a different Bench decided Sukdeb Saha v. The State of Andhra Pradesh (2025 INSC 893) — a case brought by a father who disputed the finding that his seventeen-year-old daughter, a NEET aspirant at a coaching institute in Visakhapatnam, had died by suicide. The Court transferred the investigation to the CBI and issued fifteen guidelines.
Some are addressed to coaching institutes specifically. Others are drafted for educational institutions generally: a uniform mental health policy; at least one qualified counsellor, psychologist or social worker in any institution with more than a hundred students; suicide helpline numbers displayed where students actually are; twice-yearly staff training in psychological first aid; tamper-proof ceiling fans and restricted access to rooftops.
How the two judgments fit together is stated by the Court itself. The January 2026 judgment records that the task force "has also taken due note of the elaborate guidelines issued by this Court in Sukdeb Saha", and describes its own nine directions as "over and above the guidelines which have been issued" there. The obligations accumulate. Neither replaces the other, and an institution answerable to one is answerable to both.
Which of these rules actually bind an IIT
An IIT is not an ordinary university. It takes its degree-granting power from the Institutes of Technology Act, 1961 and is an Institute of National Importance — a category the University Grants Commission Act treats separately from the universities and colleges its regulations are written around. The distinction is not academic. It decides who an institution answers to.
The January 2026 judgment uses that distinction directly. Direction (iv) requires the annual report of student suicides and unnatural deaths to go to the UGC and the professional regulators; then, in the same breath, it separates out Central Universities and Institutes of National Importance, whose reports "must be reported to the Department of Higher Education, Ministry of Education, Government of India." For the IITs, the reporting line runs to the Ministry.
Direction (ix) is often read as saying the opposite — that every institution must comply with the four named UGC regulations on ragging, equity, sexual harassment and grievance redressal. Its words are narrower. Institutions are put to strict notice "to remain fully compliant with all the regulations that have a binding effect on them", and the four are then listed as examples. The Court required compliance with whatever binds an institution. It did not decide which of those four bind an Institute of National Importance, and this page does not decide it either.
In practice the answer has been inconsistent. The UGC has named IITs on a public anti-ragging non-compliance list; it approached the same institutions about grievance redressal by advisory letter. When the Ministry reviewed student deaths at one IIT in January 2026, it measured the institute against its own 2023 framework rather than against a UGC regulation.
Aarunya takes no position on how the question should be resolved. It should be resolved, in writing, by the authority competent to resolve it. An institution cannot be held to a rule while no one will say whether the rule applies to it, and a citizen cannot check compliance with a rule whose application is undecided.
What we found
Of six institutional, RTI and Commission claims examined at IIT Madras and IIT Kanpur, none verified as originally framed.
One cited ruling reached the opposite conclusion from what was claimed of it.
0 of 6
aggregate claims held up as originally framed
Individual case-level reporting held up well. The figures meant to explain the scale did not.
Why this matters
The Court built the machinery. The public still cannot see inside it.
In January 2026 the Supreme Court required central maintenance of suicide data for ages 15–29, required the NCRB to separate school students from higher-education students, required institutions to report student suicides and unnatural deaths to police immediately, and required annual reports to the UGC, the professional regulators, or — for the IITs and other Institutes of National Importance — the Department of Higher Education. What no direction requires is that any of this reach the public in a checkable form. The reports go to a regulator or a ministry, not to citizens. There is no shared definition, no shared classification, no case identifier and no independent audit. The machinery exists. The window does not.
A number nobody can check cannot hold anyone to account
Parliament has been given all-IIT totals. RTI applicants have been given institute-level counts. Independent compilers have published far larger cumulative figures. Each may be accurate on its own terms, but they cover different periods, different institutions and different definitions, and the documents behind almost all of them cannot be retrieved by a member of the public. That is the accountability failure: not that a figure is too high or too low, but that nobody outside the system can test it.
We tested the record, and the official-looking figures held up worst
Across three deep institute audits, individual deaths reported case by case held up well under checking. The aggregate claims meant to explain the scale did not. Of six institutional, RTI and Commission claims examined at IIT Madras and IIT Kanpur, none verified as originally framed. One cited Commission ruling reached the opposite conclusion from what was claimed of it — it upheld non-disclosure. The one clean reconciliation achieved anywhere worked precisely because its period, institute and definition all lined up.
This is fixable under existing powers
Everything proposed here can be implemented by regulators, the Ministry and the institutions themselves. It needs no new primary legislation and no fresh litigation. It asks for counts, classifications and traceability — not the personal details of any student, which a Commission proceeding in 2026 declined to order disclosed on privacy grounds, and which we are not asking for either.
The orders have accumulated. The record of whether they were followed has not.
Two Supreme Court judgments now govern this ground. In July 2025, Sukdeb Saha v. The State of Andhra Pradesh issued fifteen guidelines to educational institutions. In January 2026, Amit Kumar added nine directions the Court described as over and above them. Between them they require mental health policies, counsellors, helplines, staff training, immediate police reporting of every student death, annual reports to a regulator or to the Ministry, medical care within a kilometre, posts filled within four months and scholarship backlogs cleared. What no part of either judgment requires is that the resulting record be published. Eight months after the second judgment, for most of the nine directions there is no public evidence either way — which is a statement about what can be seen, not a finding that nothing was done. That gap between what is required and what is checkable is the subject of this issue.
What we are asking for
- 1NATIONAL REFORM 1 · Adopt a single definition of "student" for death reporting across all higher education, consistent with the Supreme Court's direction that reporting cover classroom, distance and online students alike, with a companion register recording every other campus death separately. Done when a regulator has published the definition and institutions report against it.
- 2NATIONAL REFORM 2 · Adopt one classification vocabulary across all institutions and regulators, distinguishing confirmed, officially classified, suspected, undetermined, accidental, natural and homicide cases, and require every published total to state which categories it includes. Done when a published national total names its categories.
- 3TRANSPARENCY 3 · Require institute totals and national totals to reconcile for each year, with any variance published and explained. Done when a national figure is published alongside the institute figures that sum to it.
- 4TRANSPARENCY 4 · Build the reporting channel, then publish what it collects. The Supreme Court requires an annual report of student suicides and unnatural deaths to reach the UGC, the professional regulators, or — for the IITs and other Institutes of National Importance — the Department of Higher Education. No public circular, notification or portal establishing that channel for Institutes of National Importance could be located for this review. Aarunya asks first that the Department publish the mechanism itself — the format, the deadline and which institutions have filed — and second that the resulting reports be published institute-wise, year-wise and machine-readable. Done when the mechanism is public and the reports are downloadable without an RTI application.
- 5TRANSPARENCY 5 · Assign a permanent case identifier at first report, carried unchanged through institute, ministry and national records. Done when a citizen can trace one case identifier across all three.
- 6TRANSPARENCY 6 · Publish a correction history whenever a figure changes, showing the previous value, the new value, the date and the reason. Done when a revised figure is published with its prior value still visible.
- 7TRANSPARENCY 7 · Commission an independent annual audit of a sample of records against their underlying documents, by a body that does not report to the institutions audited, published in full including failures. Done when the first such audit is published.
- 8IIT REFORM 8 · Every IIT publishes its own student-death register for the last ten years on its own website, using the standard above — in addition to the annual report it already owes the Department of Higher Education. Done when all 23 registers are online.
- 9IIT REFORM 9 · Where an institution has previously given a figure to Parliament, a Commission or an RTI applicant, it publishes the underlying record — or states plainly that it cannot, and why. Done when each previously published figure carries either its record or that statement.
- 10STUDENT SUPPORT 10 · Publish compliance data against each of the Supreme Court's institutional-condition directions separately, on a fixed annual cadence — round-the-clock medical access within one kilometre, vacant teaching and non-teaching posts filled within four months, key administrative posts filled within four months, scholarship backlogs cleared within four months. Not a compliance statement but the underlying figures, institution by institution, so that a reader can tell a met obligation from an unmet one. Done when each direction has its own published data column per institution, dated.
- 11STUDENT SUPPORT 11 · Verify mental-health infrastructure claims independently before publication, stating the date and the population each figure describes. In our audits most such figures could not be located, or described a different population or a different institute. Done when published counsellor and helpline figures carry a verification date and a defined population.
- 12ACCOUNTABILITY 12 · Publish the task force's own output in full. The interim report submitted to the Supreme Court on 6 November 2025, and the final report when it is filed, should be published as documents rather than summarised through press releases — together with institution-level survey response data showing which institutions answered and which did not. The Court itself recorded a 3.5 per cent institutional response rate to a survey the Ministry calls mandatory. That is a matter of public record and should be visible as one.
- 13ACCOUNTABILITY 13 · Settle, publicly, which rules bind an Institute of National Importance. The January 2026 judgment routes death reporting for the IITs to the Department of Higher Education, and separately requires compliance with the regulations that have a binding effect on each institution — without deciding which of the four named UGC regulations bind an Institute of National Importance. In practice the UGC has enforced its anti-ragging regulation against IITs by name while approaching grievance redressal by advisory letter. Aarunya takes no position on the answer. It asks that the authority competent to give one does so in writing, because an institution cannot be held to a rule while no one will say whether the rule applies to it.
- 14ACCOUNTABILITY 14 · State, once a year, what has actually been done about each direction. For most of the nine directions issued in January 2026, this review could locate no public evidence in either direction — neither of compliance nor of failure. An annual statement from the Department of Higher Education recording, for each direction, what was required, what was done and what remains outstanding would replace that silence with a record. Done when such a statement exists and is published.
Facts & sources
Every number here is checkable
We link the source for each claim and record the date we last verified it. If you find something wrong, tell us and we will correct it in public.
On 24 March 2025, in Amit Kumar & Ors. v. Union of India, Criminal Appeal No. 1425 of 2025, the Supreme Court recorded the deaths of two IIT Delhi students on 08.07.2023 and 01.09.2023, set aside the High Court's refusal to order registration of a criminal complaint, and held that information disclosing a cognizable offence must be registered as an FIR under section 154.
Supreme Court of India, 2025 INSC 384 · court record · verified 2026-08-23
Verified
The official PDF at this URL could not be parsed on retrieval. The judgment's content is confirmed against a full reproduction of its text; the URL is published because it is the official location of the document.
On 15 January 2026, exercising its powers under Article 142, the Supreme Court issued nine directions (2026 INSC 62). Four concern data: Sample Registration System suicide data for ages 15–29 centrally maintained; the NCRB to distinguish school-going students from higher-education students; every institution to report any student suicide or unnatural death to police as soon as it knows, regardless of where the death occurred and whether the student studied in classroom, distance or online mode; and an annual report of such deaths to the UGC and professional regulators, or — for Central Universities and Institutes of National Importance, which include the IITs — to the Department of Higher Education, Ministry of Education. Five concern conditions: round-the-clock medical help on campus or within one kilometre; all vacant faculty positions, both teaching and non-teaching, filled within four months with priority to reserved posts; key administrative posts filled within four months with annual reporting on reserved-post vacancies; scholarship backlogs cleared within four months, with no student barred from examinations, hostel or class for a disbursement delay; and full compliance with the UGC regulations of 2009, 2012, 2016 and 2023.
Supreme Court of India, 2026 INSC 62, paragraph 45 · court record · verified 2026-08-23
Verified
Read from the official Supreme Court PDF — 38 pages, paragraph 45 quoted directly.
None of the nine directions requires the resulting record to be published where the public can check it. The directions require reporting to the police, to regulators and to the Ministry. They establish no shared definition of "student", no shared classification vocabulary, no case identifier, no obligation for institute and national totals to reconcile, no correction history, and no independent audit.
Supreme Court of India, 2026 INSC 62, paragraph 45 · court record · verified 2026-08-23
Verified
Established by reading the full 38-page judgment and searching it for publication language: the only occurrence of "website" refers to the National Task Force's own survey site, and no direction in paragraph 45 requires public disclosure. This is a statement about this judgment, not a claim that no such requirement exists anywhere else in Indian law.
Aarunya's own audit reports cite two different all-IIT student-suicide figures for the same stated period of 2018–2023: 39, from an answer the Delhi report dates 27 July 2023, and 33, from an answer the Kanpur report dates 16 March 2023. Both are described as IIT-only for that period. The audit does not establish why the two figures differ, and Aarunya selects neither as correct.
Aarunya IIT student death audit, Delhi report section 6 and Kanpur report section E · parliamentary record · verified 2026-08-27
Primary source not located
Neither parliamentary answer was retrieved as an official document; both reach the audit through secondary reporting. Aarunya presents neither as a settled national figure. A provenance conflict is now recorded: Aarunya's own report attributes the figure of 39 to a parliamentary answer it dates 27 July 2023, while later secondary reporting attributes 39 to data placed before the Supreme Court. These may refer to the same underlying figure at different stages, but that has not been established and Aarunya does not resolve it.
Aarunya's Delhi and Kanpur reports also differ on which parliamentary answer the combined figure of 61 for IITs, NITs and IIMs belongs to — the Delhi report associates it with the answer it dates 27 July 2023, the Kanpur report with the answer it dates 16 March 2023.
Aarunya IIT student death audit, Delhi report section 6 and Kanpur report section E · parliamentary record · verified 2026-08-27
Unresolved
Recorded rather than resolved. Aarunya does not determine which attribution is correct.
A parliamentary answer dated 20 December 2021 reported 34 student suicides across all IITs for 2014–2021, with a category breakdown of SC 5, ST 0, OBC 13 and Minority 0. It is an all-IIT aggregate and cannot be mapped to any single institute.
Aarunya IIT student death audit, Kanpur report section E · parliamentary record · verified 2026-08-27
Primary source not located
The original answer was not retrieved; it reaches the audit through secondary reporting.
An independent compilation reported a cumulative pan-IIT figure of 115 for 2005–2024, later restated at approximately 160. It combined RTI responses from only 13 of the 23 IITs with National Crime Records Bureau statistics and parliamentary answers to fill the gaps, and the total moved without a republished institute-by-institute breakdown.
Aarunya IIT student death audit, Madras report sections D and E, Kanpur report sections E and G · Aarunya audit · verified 2026-08-27
Unresolved
An attributed claim by a named compiler, not audited data. Aarunya's own reports differ on the date of the later restatement — one records March 2026, the other August 2026 — and that conflict is not resolved here.
Figures of 26 for IIT Madras and 18 for IIT Kanpur, both for 2005–2024, are single lines of that same compilation rather than verified institute totals.
Aarunya IIT student death audit, Madras report section E and Kanpur report sections E and G · Aarunya audit · verified 2026-08-27
Unresolved
Neither is a clean single-institute RTI figure. Aarunya does not publish either as an established institute total.
Four second appeals concerning IIT Madras, IIT Jodhpur, IIT Goa and IIT Kanpur were heard together by the Central Information Commission on 15 April 2026 and decided on 30 April 2026. The appeals sought details of students who had died by suicide or unnatural death. IIT Madras disclosed aggregate death numbers by year for 2019 to 2025; IIT Kanpur disclosed gender, academic programme, date, native state and location; IIT Jodhpur reported nil cases; IIT Goa withheld information as third-party personal data. Names, age and caste were withheld throughout under section 8(1)(j) of the Right to Information Act, and the Commission upheld that withholding. The Commission recorded a persistent crisis of multiple suicides annually across campuses, with high concentration at IIT Kanpur and IIT Kharagpur, and recommended high-level committees together with proactive disclosure of information about those committees.
Central Information Commission, decision of 30 April 2026 in CIC/IITMD/A/2025/634151, CIC/IITJP/A/2025/634132, CIC/IITGO/A/2025/634131 and CIC/IITKP/A/2025/636407 · CIC order · verified 2026-09-02
Verified
Read from the order itself. It establishes what each institute disclosed or withheld in these four proceedings, the section relied on, and the Commission's own observations and recommendations. It does not establish a national IIT suicide total, does not verify Aarunya's cross-institute count of 29, and is not evidence for any individual death within that count. Names, age and caste remain withheld, and Aarunya does not seek them.
Research inherited by the audit cited a Commission case under a name that could not be located in the Commission's searchable database at the time of the audit, and characterised it as authority that IIT suicide data had been ordered disclosed. The Commission's decision of 30 April 2026 cites that earlier matter as CIC/DS/A/2011/004488/RM, dated 1 February 2013, and records that it directed the institute to provide only the number of students who had died by suicide, expressly holding that personal information need not be provided. The inherited characterisation was therefore right that aggregate numbers had been ordered disclosed, and wrong to the extent it implied personal details had been. Aarunya's own earlier correction, which described the adjacent case as having upheld non-disclosure outright, was itself too broad and is corrected here.
Aarunya IIT student death audit, Kanpur report section G · CIC order · verified 2026-09-02
Corrected
Corrected against the primary source. The earlier matter is cited within the 30 April 2026 order rather than retrieved as a standalone document, so its case number and holding are taken from that order. Aarunya does not publish the name under which the citation was originally inherited.
Across three completed deep institute audits — IIT Delhi, IIT Madras and IIT Kanpur — 29 cases are classified as suicide or suspected suicide: 9, 11 and 9 respectively, from 49 leads examined and 42 deaths of all affiliations documented. This is NOT a national IIT total; it covers 3 of India's 23 IITs.
Aarunya IIT student death audit, overall report · Aarunya audit · verified 2026-08-27
A floor rather than a ceiling: it reflects what could be independently verified, so a death never reported would not appear in it. The remaining 20 IITs have not been examined and no figure is published for them.
Of the six institutional, RTI and Commission-level claims examined across IIT Madras and IIT Kanpur, none verified as originally framed.
Aarunya IIT student death audit, overall report section 4 · Aarunya audit · verified 2026-08-27
The summary of the overall report states this as nine claims across those two institutes. Its own master table gives six, with nine being the total across all three institutes including Delhi, where two claims did verify. The controlling figure is 0 of 6.
No FIR text, post-mortem report, inquest finding, court order, RTI reply or Commission order was retrieved in full for any individual death at any of the three institutes audited. Every individual classification rests on contemporary journalism, with two exceptions where a genuine institutional document was located.
Aarunya IIT student death audit, overall report section 4 · Aarunya audit · verified 2026-08-27
Primary source not located
The largest and most consistent evidence gap in the programme, and the reason all three institutes stand at conditional pass rather than fully verified.
IIT Delhi's 9 comprise 6 cases classified as suicide by police or the institution and 3 classified as suspected or provisional. No IIT Delhi case in the audit reaches the standard of a judicially established finding.
Aarunya IIT student death audit, Delhi report section 6 · Aarunya audit · verified 2026-08-23
The third provisional IIT Delhi case is a postgraduate student who died in August 2026. As of the audit there was no institute statement and no police classification, the case was one day old, and it was expressly not upgraded on the basis of media word choice.
Aarunya IIT student death audit, Delhi report sections 3 and 7 · Aarunya audit · verified 2026-08-23
Provisional
RE-CHECK REQUIRED — this case remains provisional and must be re-checked, per the audit. No institute statement and no police classification. Counted in the 9 and excluded from the Delhi report's own closing count of 8 — see corrections. NOT described as a confirmed suicide.
An RTI reply reported 6 suicides at IIT Delhi across academic years 2019-20 to 2023-24. Counting the police-classified and provisional cases — codes B and C — in the audit's independently built Delhi ledger by Indian academic year over the same window also gives exactly 6. This is the only clean reconciliation between an official figure and an independent count anywhere in the programme.
Aarunya IIT student death audit, Delhi report sections 2 and 6 · RTI reply · verified 2026-08-23
Primary source not located
The RTI reply itself was not retrieved; the figure reaches the audit through contemporary reporting of the reply. The 6 matched are specifically the B and C cases in that window, not all documented deaths.
A published account described a March 2026 IIT Delhi death as the "eighth suicide since 2021." The audit's ledger finds six by that date. The two-case gap is unexplained and the underlying data is not cited in the source. Aarunya does not publish "8 since 2021" as a verified figure.
Aarunya IIT student death audit, Delhi report sections 2 and 6 · Aarunya audit · verified 2026-08-23
Unresolved
At IIT Madras, 19 leads were examined, 11 individual student deaths met the threshold for inclusion, and 6 were correctly excluded — three postdoctoral scholars, one faculty member, one faculty spouse with no institutional affiliation of her own, and one student of a different institution resident on campus.
Aarunya IIT student death audit, Madras report sections B and C · Aarunya audit · verified 2026-08-27
At IIT Kanpur, 17 leads were examined, 9 individual student deaths met the threshold for inclusion, and 4 were correctly excluded — a security guard, an assistant professor, an assistant registrar and a postdoctoral researcher. Two further institutional claims — a statement of "2 suicides since 2014" and an RTI figure of 17 deaths for 2005–2025 — matched no locatable document.
Aarunya IIT student death audit, Kanpur report sections B, C and E · Aarunya audit · verified 2026-08-27
Primary source not located
The exclusion of the assistant professor was confirmed against the institute's own departmental memorial page — one of only two genuine institutional documents located in the entire programme. Aarunya publishes neither the "2 since 2014" nor the "17" figure.
Independent verification found specific, material errors in the research the audit started from. An allegation about academic backlogs at IIT Kanpur was traced to an unrelated student who shared a surname. A claim that an IIT Madras supervisor had been restricted from a lab after a death inverted the true, opposite fact. A cited count of "34 recommendations" matched no located document. Ten of ten flagged non-student exclusions were, by contrast, confirmed correct.
Aarunya IIT student death audit, overall report section 4 · Aarunya audit · verified 2026-08-27
A parliamentary answer dated 2 December 2019 reported 50 student suicides across the IITs over the preceding five years, and the reporting states the deaths occurred at 12 of the 23 IITs. The institute-wise figures carried in that reporting name IIT Guwahati at 14, IIT Bombay at 7 and IIT Madras at 7, with the remaining 22 spread across nine other IITs that the reporting does not name.
Parliamentary answer of 2 December 2019, as reported in contemporary journalism · parliamentary record · verified 2026-09-02
Primary source not located
PRIMARY SOURCE NOT LOCATED — the parliamentary answer itself was not retrieved; this reaches Aarunya as a reported parliamentary figure. It is published as a reported figure, not a verified one. Its period and institutional scope both differ from Aarunya's own audits, and the named breakdown accounts for three of the twelve institutes only. It is not comparable with, and must not be added to, Aarunya's 29.
An RTI response reported 27 student suicides across 10 of the IITs between 2014 and 2019. The institute figures carried in that reporting include IIT Madras at 7, IIT Kharagpur at 5, and IIT Delhi and IIT Hyderabad at 3 each.
RTI response, as reported by The Statesman · RTI reply · verified 2026-09-02
Primary source not located
PRIMARY SOURCE NOT LOCATED — the RTI response itself was not retrieved; this is a secondary report of an RTI reply. Its scope covers 10 of the 23 IITs over a different period from Aarunya's audits, and without the underlying records it is not comparable with Aarunya's 29 and must not be added to it.
Direction (iv) of the January 2026 judgment separates the Institutes of National Importance from the ordinary regulatory channel. An annual report of student suicides or unnatural deaths must go to the UGC and the professional regulators; but "In case of Central Universities and Institutes of National Importance (for short, the 'INIs'), or any HEI that does not fall within the above-mentioned framework, it must be reported to the Department of Higher Education, Ministry of Education, Government of India." For the IITs, the reporting line runs to the Ministry.
Supreme Court of India, 2026 INSC 62, paragraph 45(iv) · court record · verified 2026-09-11
Verified
Quoted from the official judgment PDF, read directly at page 28.
Direction (ix) is narrower than it is often read to be. It puts all higher education institutions to strict notice "to remain fully compliant with all the regulations that have a binding effect on them", and then lists the UGC regulations on ragging (2009), equity (2012), sexual harassment (2016) and grievance redressal (2023) as examples. The Court required compliance with whatever binds an institution. It did not decide which of those four bind an Institute of National Importance.
Supreme Court of India, 2026 INSC 62, paragraph 45(ix) · court record · verified 2026-09-11
Verified
Quoted from the official judgment PDF, read directly at pages 31-32. Whether each of the four regulations binds an Institute of National Importance is not settled by this judgment and is not asserted here.
The nine directions of January 2026 sit on top of an earlier set, not in place of them. The judgment records that the National Task Force "has also taken due note of the elaborate guidelines issued by this Court in Sukdeb Saha v. The State of Andhra Pradesh reported in 2025 SCC OnLine SC 1515", and describes its own directions as "over and above the guidelines which have been issued in the decision of this Court in Sukdeb Saha".
Supreme Court of India, 2026 INSC 62, paragraphs 12 and 44 · court record · verified 2026-09-11
Verified
Sukdeb Saha v. The State of Andhra Pradesh was decided on 25 July 2025 by Justices Vikram Nath and Sandeep Mehta (2025 INSC 893). It arose from the death of a 17-year-old NEET aspirant at a coaching institute in Visakhapatnam whose father disputed the finding of suicide. The Court transferred the investigation to the CBI and issued fifteen guidelines. Some are addressed to coaching institutes specifically; others are drafted for educational institutions generally, including a uniform mental health policy, at least one qualified counsellor, psychologist or social worker in institutions with more than a hundred students, displayed helpline numbers, twice-yearly staff training in psychological first aid, and tamper-proof ceiling fans with restricted rooftop access.
Supreme Court of India, Sukdeb Saha v. The State of Andhra Pradesh, 2025 INSC 893 · court record · verified 2026-09-11
Verified
Read against a full reproduction of the judgment text rather than the Court's own PDF. The guideline numbering and the split between coaching-specific and generally-drafted guidelines should be confirmed against the official text before any individual guideline is quoted as binding on a particular institution.
The Court recorded how thin the evidence beneath its own interim findings was. Of 60,383 higher education institutions, the response rate to the National Task Force survey was "relatively low i.e., 3.5%" — which the Court nonetheless treated as giving "a good representative idea of the reality in HEIs in this country", supplemented by 30 meetings at 19 institutional sites.
Supreme Court of India, 2026 INSC 62, page 7 · court record · verified 2026-09-11
Verified
The Court treated its own nine directions as unfinished business, not as the conclusion of the process: they "would be in addition to the existing work of the NTF and what they plan to incorporate in their final report", and the Court said it hoped the Task Force would propose a model 'Universal Design Framework', 'Suicide Prevention and Postvention Protocol' or 'Student well-being Protocol'.
Supreme Court of India, 2026 INSC 62, paragraphs 48 and 49 · court record · verified 2026-09-11
Verified
The National Task Force on Student Mental Health and Suicide Prevention in Higher Educational Institutions was constituted by the Supreme Court and is chaired by former Justice Ravindra Bhat. Its remit is not confined to any one institution or category of institution: it brings all types of higher education institution across the country within its purview. It launched a public website to collect stakeholder input on 8 August 2025.
Press Information Bureau, Ministry of Education, 8 August 2025 · government document · verified 2026-09-11
Verified
The Task Force submitted an interim report to the Supreme Court on 6 November 2025 and, in the Ministry's own words, "awaits cognizance from the Court". The Ministry describes the institutional survey behind it as mandatory: "It is mandatory for HEIs to fill out the institute survey", covering faculty diversity, vacant positions, student composition and the provision of mental health services.
Press Information Bureau, Ministry of Education, 6 November 2025 · government document · verified 2026-09-11
Verified
The Task Force's final report had not been filed when this page was last verified. The deadline has been reported as 31 October 2026, extended by the Supreme Court on 27 May 2026.
Supreme Court order of 27 May 2026, as reported in contemporary journalism · media report · verified 2026-09-11
Primary source not located
The extension itself is consistently reported across several outlets but the order text was not retrieved. What is verified from the judgment is only that a final report was still expected. Treat the date as reported, not as read.
The regulatory track Direction (ix) points at moved backwards during the compliance window. The UGC (Promotion of Equity in Higher Educational Institutions) Regulations, 2026 were stayed by the Supreme Court on 29 January 2026, days after notification, the Bench holding that some provisions "suffer from certain ambiguities, and the possibility of their misuse cannot be ruled out" and directing that the 2012 Regulations "will continue to operate and remain in force till further orders". The matter was referred to a larger Bench.
Supreme Court of India, order of 29 January 2026, as reported by SCC OnLine · court record · verified 2026-09-11
Primary source not located
The order text itself was not retrieved. The 2012 Regulations, not the 2026 ones, are the operative equity regulations.
No publicly indexed circular, notification or portal establishing the reporting channel Direction (iv) requires — the route by which an Institute of National Importance files its annual report with the Department of Higher Education — was located for this review.
Aarunya review of published government material, September 2026 · Aarunya audit · verified 2026-09-11
Primary source not located
This is a statement about what could be found in public, not a finding that no mechanism exists. An internal instruction could exist without public indexing. The question is answerable by a single RTI application to the Department of Higher Education, which is the next step in this programme.
The UGC has enforced its anti-ragging regulation against IITs by name notwithstanding the jurisdictional question: in mid-2025 it listed 89 institutions as non-compliant for failing to file mandatory anti-ragging undertakings, among them four IITs. Whether those institutions subsequently came into compliance was not established for this review.
UGC anti-ragging non-compliance list, mid-2025, as reported in contemporary journalism · media report · verified 2026-09-11
Primary source not located
Reported consistently across outlets; the UGC list itself was not retrieved. Named here to show how the regulator has acted in practice, not to characterise any institution's current compliance.
What is still open
The January 2026 directions were not the end of the Court's own process, and reading them as a finished settlement is the most common mistake made about this case. The Court issued them on the strength of an interim report and said plainly that more was expected: the directions "would be in addition to the existing work of the NTF and what they plan to incorporate in their final report."
The National Task Force on Student Mental Health and Suicide Prevention in Higher Educational Institutions was created by the same order of 24 March 2025 that required FIRs in the two IIT Delhi deaths. It is chaired by former Justice Ravindra Bhat, runs out of the Ministry of Education, and covers every kind of higher education institution rather than any single category. It filed an interim report with the Supreme Court on 6 November 2025 and, in the Ministry's own words, "awaits cognizance from the Court". Its final report — the document the Court hoped would carry a model suicide prevention and postvention protocol — had not been filed when this page was last verified. The deadline has been reported as 31 October 2026.
The Court also recorded how much evidence that interim work rested on. Of 60,383 higher education institutions, the response rate to the task force's survey was "relatively low i.e., 3.5%" — a survey the Ministry describes as mandatory for institutions to complete. The Court treated what came back as giving "a good representative idea", and supplemented it with thirty meetings at nineteen institutional sites. Both halves belong in the same sentence: the Court has acted, and by its own account the evidentiary base beneath the action is thin.
Two things follow, and they are different in kind. The first is that the consolidating mechanism this judgment relies on is unfinished — that is a fact, and a dated one. The second is that for most of the nine directions, no public evidence of implementation could be located either way. That is not a finding that anything was ignored. It is a description of what a citizen can see from outside, which is the subject of this issue.
What public evidence shows, direction by direction
| Direction | What public evidence shows | Basis |
|---|---|---|
| i · Suicide data for ages 15-29 centrally maintained | No public evidence located | No publication located either way |
| ii · NCRB to separate school from higher-education student suicides | No public evidence located | No post-judgment edition carrying the split was located |
| iii · Every student death reported to the police immediately | Partial, and tested once | One institute adopted a formal protocol and used it in a live case in March 2026; adoption across institutions was not established |
| iv · Annual report to the regulator; for Institutes of National Importance, to the Department of Higher Education | Mechanism not publicly located | No circular, notification or portal establishing the channel was found |
| v · Round-the-clock medical help on campus or within one kilometre | No public evidence located | A single institutional account, which is not a compliance record |
| vi · Vacant teaching and non-teaching posts filled within four months | No public evidence located | A parliamentary reply on posts filled spans 2022-2026 and does not reference the judgment or its deadline |
| vii · Key administrative posts filled within four months, reported annually | No public evidence located | Routine recruitment notices only |
| viii · Scholarship backlogs cleared within four months | No public evidence located | A related fellowship scheme was reported still delayed inside the window; not institution-specific and not conclusive |
| ix · Compliance with the regulations that bind each institution | Mixed | The 2026 equity regulations were stayed on 29 January 2026 and the 2012 regulations restored; the anti-ragging regulation has been enforced against IITs by name |
| The task force final report and model protocol | Not yet filed | Interim report 6 November 2025; final report reported due 31 October 2026 |
"No public evidence located" means exactly that: this review could not find published evidence in either direction. It is not a finding that a direction was ignored, and it should not be read as one — treating absent evidence as proof of failure is the same error this issue objects to when it is made about numbers. Several of these rows could be closed by a single RTI application, which is the nex
Aarunya's proposed Student Death Reporting Standard
None of what follows has been ordered by any court. It is Aarunya's proposed reform, built on the classification discipline we tested across three institute audits. The Court has required this information to be collected and reported. We propose making the resulting record publicly checkable.
D1 · One definition of "student", and a companion register. Primary register: persons enrolled in a degree or diploma programme on the date of death, across every mode of enrolment — which aligns with the Court's own direction that reporting cover classroom, distance and online students alike. Companion register: every other death on institutional premises or of institutionally affiliated persons — postdoctoral researchers, faculty, administrative and contract staff, security personnel, family members in institutional housing, students of other institutions resident on campus — recorded separately, never merged into the student count. In our audits, ten of ten flagged non-student cases were correctly excluded on checking, which is why both errors need preventing.
D2 · A classification vocabulary that grades evidence rather than guessing outcomes. Aarunya's proposed vocabulary: A — confirmed suicide, finally established, for example by a judicial finding. B — police or institutionally classified suicide. C — suspected or provisional suicide. D — undetermined death. E — accidental death. F — natural or medical death. G — homicide investigation open. H — judicially established homicide. Two operating rules: a classification may be upgraded only on new official evidence, never because media wording changed; and every published total must state which codes it includes, because "suicide" alone is ambiguous while "A and B" is not.
D3 · A permanent case identifier. Institute–year–sequence, assigned at first report, never reused, never renumbered — so the same case can be followed from the institute register to the Ministry's annual report to any national total.
D4 · The record itself. One row per death, with every field carrying a value or an explicit reason it is unavailable. Case identifier; institute; academic and calendar year; register; programme level only, never programme name; classification code; whether confirmed, provisional or unresolved; whether police were informed and when; FIR status; investigation status; external inquiry status; final classification or "pending"; source documents each marked public or withheld with a reason; last verified date; version number; correction log. "Not recorded" and "pending" are legitimate published values. A zero standing in for "we did not look" is not.
D5 · Publication, institute-wise and year-wise. Each institution publishes its register annually on a fixed date in machine-readable form on its own website. The recipient — regulator or Ministry — publishes the aggregate institute by institute and year by year, never as a single national number without the breakdown beneath it. Historical years are republished each time, so a changed past figure is visible rather than quietly replaced.
D6 · Reconciliation, made mandatory. The sum of institute registers must equal the published national total for the same year, register and code set; where it does not, the variance is published with its reason. This is the direct answer to two ministerial answers giving different all-IIT figures for the same stated period with nothing published to explain why.
D7 · Privacy by default. No names. No programme name, department, hostel or any field identifying an individual within a small cohort. No cause-of-death narrative, no personal circumstances, no medical history. Category data such as caste, gender or disability published only in aggregate at national level, never per case. A family may ask for a name to be associated with a case; the default is never. This is deliberately built to survive the privacy objection that succeeded before the Central Information Commission in 2026 — it asks for counts, classifications and traceability, none of which requires identifying anyone.
D8 · An independent annual audit. A sample of records checked against their underlying documents each year, by a body that does not report to the institutions being audited, published in full including failures. Each institution receives a status — verified, conditional, or not verified — on the same evidence standard we applied to ourselves.
What a citizen should be able to see
Pick an institute. Pick a year. See what is known and how well it is known, without filing an RTI application. Each row shows institute, year, register, classification code, number of cases, whether confirmed, provisional or unresolved, a link to the underlying document, the date last verified, and how many corrections have been made.
Four rules separate a dashboard from a press release. Every cell traces to a document or states why it cannot. "Not audited" and "not reported" are visible states, never rendered as zero. Codes are never silently merged — a total may combine categories, but must say which and show the split. Corrections are a column, not a footnote.
…must be centrally maintained for better and more accurate estimates.
39 vs 33
two official all-IIT totals for the same stated period, 2018–2023
Both from parliamentary answers. Aarunya selects neither as correct.
Questions
What exactly is Aarunya asking for?+
That the student-death data institutions are already required to collect and report be published in a form a citizen can check — a shared definition, a shared classification, a case identifier, reconciliation between institute and national totals, a correction history and an independent audit.
Hasn't the Supreme Court already fixed this?+
It has fixed a large part. In January 2026 it required central data maintenance, NCRB disaggregation, immediate police reporting, and annual reports to the UGC, the professional regulators, or — for the IITs and other Institutes of National Importance — the Department of Higher Education. What it did not require is that any of this be published where the public can check it. That gap is what this issue addresses.
Is the A–H classification a legal standard?+
No. It is Aarunya's proposed vocabulary, developed and tested across three institute audits. No court or regulator has adopted it, and we do not claim otherwise.
What is the national number of student suicides in the IITs?+
We do not know, and on the current record neither does anyone who cannot obtain the underlying documents. Several figures exist for different periods and institution sets. We have not selected one as the truth; establishing that number is what we are asking for.
What does the "29" mean?+
29 cases classified as suicide or suspected suicide in the three completed institute audits — IIT Delhi, IIT Madras and IIT Kanpur. This is NOT a national IIT total. It covers 3 of India's 23 IITs and is a floor even for those three, because it reflects only what could be independently verified.
Does "not audited" mean an IIT has had no student deaths?+
No. It means we have not examined it. We publish no figure, not even zero, for an institute we have not audited.
Why can't the published figures be added together?+
Because they count different things over different periods for different sets of institutions. Adding them produces a number with no definition attached to it.
Two official figures for the same period differ. Which is right?+
We do not know. Our reports record 39 from an answer dated 27 July 2023 and 33 from one dated 16 March 2023, both described as IIT-only for 2018–2023. The audit does not establish why they differ, and we select neither. That is precisely the problem this issue is about.
Why do some primary documents remain unavailable?+
Obtaining them requires filing RTIs and retrieving court and Commission records directly — a separate phase from verifying leads. We label every affected claim rather than presenting a news report as though it were the document.
Why don't you publish the names?+
Because we are asking institutions to publish counts, classifications and traceability without identifying anyone, and we hold ourselves to the same rule. A public register does not need a name to be verifiable.
Is Aarunya claiming institutions deliberately falsified figures?+
No. We found no evidence of falsification or concealment and do not allege it. We found a record that cannot be checked. Those are different claims and we will not blur them.
What would cause Aarunya to correct this issue?+
Any primary document that contradicts a claim here, any case wrongly included or excluded, any date we got wrong. We have already corrected our own reports on this page and will do so again, dated, with the original wording retained.
What is the National Task Force, and has it reported?+
It is the expert body the Supreme Court created by the same order of 24 March 2025 that required FIRs in the two IIT Delhi deaths — formally the National Task Force on Student Mental Health and Suicide Prevention in Higher Educational Institutions, chaired by former Justice Ravindra Bhat and run out of the Ministry of Education. It covers every kind of higher education institution. It submitted an interim report on 6 November 2025. Its final report, which the January 2026 judgment expected to carry a model prevention and postvention protocol, had not been filed when this page was last verified; the deadline has been reported as 31 October 2026.
Do the Supreme Court's directions bind the IITs?+
The directions themselves do. The judgment applies to higher education institutions and deals with the Institutes of National Importance expressly, routing their annual death reporting to the Department of Higher Education rather than to the UGC. What the judgment does not decide is which of the four named UGC regulations — on ragging, equity, sexual harassment and grievance redressal — bind an Institute of National Importance. Direction (ix) requires compliance with the regulations that have a binding effect on an institution, and leaves the prior question of which those are to the underlying law. This page records that question as open rather than answering it.
Has anything actually changed since January 2026?+
For one direction there is public evidence of partial implementation. For most of the rest, this review could locate no public evidence either way. That is a statement about what can be seen from outside, not a finding that nothing was done — and the difference matters, because treating an absence of published evidence as proof of failure is the same error this issue objects to when institutions do it with numbers. What can be said with confidence is that the consolidating mechanism the judgment relied on, the task force's final report, had not been filed.
Is Aarunya saying institutions caused these deaths?+
No. Nothing on this page attributes any student's death to any policy, regulation, institution or individual. The systems described here are accountability and support systems, not established causes. Where a source's own framing implies a causal link, that framing belongs to the source and is not adopted here.
Where this stands
The whole process, in public
We publish each step as it happens — including the steps where nothing happens.
Supreme Court orders FIR registration in two IIT Delhi student deaths
24 March 2025The Court set aside the High Court's refusal and held that information disclosing a cognizable offence must be registered as an FIR.
A second Supreme Court judgment issues fifteen guidelines
25 July 2025Sukdeb Saha v. The State of Andhra Pradesh (2025 INSC 893). Brought by a father who disputed the finding that his daughter had died by suicide, the case produced fifteen guidelines to educational institutions — mental health policy, a counsellor minimum, displayed helplines, staff training, safer fittings — and a transfer of the investigation to the CBI.
The Supreme Court's task force opens its survey
8 August 2025The National Task Force on Student Mental Health and Suicide Prevention in Higher Educational Institutions, chaired by former Justice Ravindra Bhat, launches a website to collect input from students, faculty, parents and institutions. Its remit covers every kind of higher education institution, not one category.
Interim report submitted; final report still to come
6 November 2025The task force files an interim report with the Supreme Court and, in the Ministry's words, awaits cognizance from the Court. The institutional survey behind it is described by the Ministry as mandatory for institutions to complete.
Supreme Court issues nine directions on student deaths in higher education
15 January 2026Directions issued under Article 142. Four concern data: central maintenance of suicide data for ages 15-29, NCRB disaggregation, immediate police reporting wherever a death occurs and whatever the mode of study, and annual reports to the UGC, the professional regulators, or the Department of Higher Education for Institutes of National Importance such as the IITs. None requires publication.
The new equity regulations are stayed; the 2012 rules return
29 January 2026Days after notification, the Supreme Court stays the UGC equity regulations of 2026, holding that some provisions suffer from ambiguities and that misuse cannot be ruled out, and directs that the 2012 Regulations continue in force till further orders. One of the four regulations named in Direction (ix) is therefore the older one.
Information Commission upholds non-disclosure of IIT suicide demographics
~27 May 2026A proceeding against four IITs sought demographic details of students who died by suicide since 2005. Non-disclosure was upheld on privacy grounds and no figure was released.
Source: Order text not retrieved — PRIMARY SOURCE NOT LOCATED
The task force's final report deadline is extended
27 May 2026The Supreme Court is reported to have given the task force until 31 October 2026 to file its final report — the document the January judgment expected to carry a model prevention and postvention protocol. The order text was not retrieved for this review; the date is recorded as reported.
IIT Delhi audit completed — conditional pass
23 August 202613 deaths documented, 9 classified as suicide or suspected suicide. One RTI figure reconciled exactly against the independently built ledger — the only clean reconciliation in the programme so far.
IIT Madras audit completed — conditional pass
27 August 202619 leads examined, 11 verified, 6 correctly excluded. Both institutional-number claims failed to hold up as originally framed.
IIT Kanpur audit completed — conditional pass
27 August 202617 leads examined, 9 verified, 4 correctly excluded. All three institutional-data claims failed; one cited case reached the opposite conclusion from what was claimed.
Standard published; programme open
27 August 2026The proposed Student Death Reporting Standard is published for adoption. Three IITs are at conditional pass; 20 are not audited.
Final report due
31 October 2026The reported deadline for the task force's final report. Until it is filed and published, the central consolidating mechanism the January 2026 judgment relied on remains unfinished.
Next institute audit, and the primary documents
The programme continues one institute at a time. In parallel: fresh RTIs, the Commission order text, and the court documents that would move any institute from conditional pass to fully verified.
The IIT-by-IIT verification programme
A standing research programme, expanded one institute at a time. Four statuses: Not audited — not examined, no figure of any kind shown. In progress — under way, nothing published until complete. Conditional pass — case list built and corroborated, at least one gap remaining, typically no primary document retrieved. Fully verified — at least one primary document per case and every institutional claim reconciled.
IIT Delhi — Conditional pass — 9 cases (6 code B, 3 code C). What would close the gap: a fresh 2024–2026 RTI; the January 2024 High Court order; a police classification for the August 2026 case.
IIT Madras — Conditional pass — 11 cases. What would close the gap: one retrieved primary document; the Commission order text; a fresh RTI.
IIT Kanpur — Conditional pass — 9 cases. What would close the gap: one retrieved primary document; the Commission order text; the Ministry order naming a review committee; a precisely scoped RTI.
The remaining 20 IITs — Not audited. No figures shown, and none inferred.
The combined figure, stated in full: 29 cases classified as suicide or suspected suicide in the three completed institute audits — IIT Delhi, IIT Madras and IIT Kanpur. This is NOT a national IIT total. It covers three of India's 23 IITs, and is a floor even for those three, because it reflects what could be independently verified — a death never reported would not appear in it. Not all 29 are confirmed suicides: most rest on a police or institutional classification, several are explicitly provisional, and none reaches the standard of a judicially established finding.
An institute marked "Not audited" has not been examined. It has not been found to have no deaths. Nothing on this page should be added into a national figure.
Methodology, limitations and corrections
Method. Every case began as a lead, never a fact. Leads were checked against court records, government and parliamentary documents, official institute statements, police records, RTI and Commission material, and contemporary journalism — in that order. A case entered a count only when at least two independent contemporary sources agreed on the core facts. Enrolled degree-seeking students only; postdoctoral researchers, faculty, staff, family members and students of other institutions were documented and then excluded.
How to read the evidence labels. Each fact carries a label describing the state of its evidence, not the seriousness of the claim. Verified — the primary source was retrieved and independently checked. Primary source not located — a credible source reports the figure, but the underlying primary document was not retrieved. Unresolved — sources conflict, or the scope or attribution cannot presently be established. Corrected — a previously published or inherited claim was demonstrably corrected. Provisional — the case or figure remains subject to a re-check. No label — the statement describes Aarunya's own process rather than an externally verified claim. An absent label never means verified.
A–H are Aarunya's own working codes, including the "code B" and "code C" references in the verification programme above. No court, regulator, ministry or other government body has adopted these codes.
Why the public figures cannot simply be added up. Parliamentary answers, RTI responses, institutional statements and independent compilations each report figures for student deaths at the IITs, but they differ in period, institutional scope, population counted, classification and methodology, and in almost every case the underlying document cannot be retrieved by a member of the public. Aarunya's case-level audits of three IITs record 29 cases classified in those audits as suicide or suspected suicide. This is not a national total and must not be added to the separately reported figures below. Independent journalism has reached the same conclusion about the reconciliation problem; that corroborates the difficulty, and establishes no particular number.
The compiler associated with the larger independent IIT suicide-count compilation is also reported as the appellant in the CIC matter concerning IIT disclosure. This is provenance context, not independent verification of either figure.
A report referred to in the research process as the "Top 7" report could not be independently located in the project, storage, or expanded web search. The closest located material is a 2 December 2019 Lok Sabha answer containing an institute-wise breakdown. We do not assert that the two are the same document.
The central limitation. Not one FIR, post-mortem report, inquest finding, court order, RTI reply or Commission order was retrieved in full for any individual death at any of the three institutes. Every classification rests on contemporary journalism, with two exceptions where a genuine institutional document was located. All three institutes stand at conditional pass; none is fully verified.
What this does not establish. Not a national count. It does not establish why any individual student died. It finds no named person responsible for anything. It does not allege that any institute falsified or concealed information. It does not treat its own numbers as final.
Corrections to our own work. The summary of our overall report says nine institutional claims were examined across IIT Madras and IIT Kanpur and none held up. Its own master table gives six for those two institutes — nine is the total across all three including Delhi, where two claims did hold up. The controlling figure is 0 of 6, and the summary wording was wrong.
Our cross-institute table and the Delhi case reconciliation both count 9 for IIT Delhi — six code B, three code C. The Delhi report's closing certification block counts 8, excluding the third provisional case, a postgraduate student who died in August 2026 days before that audit closed. We publish 9 and name the difference. That case has no institute statement and no police classification, our own report says it must be re-checked before publication, and it is not described here as a confirmed suicide.
Our Delhi report cites 39 IIT student suicides for 2018–2023, from an answer it dates 27 July 2023. Our Kanpur report cites 33 for the same stated period, from an answer it dates 16 March 2023. The audit does not establish why the two figures differ, and Aarunya selects neither as correct. Our two reports also differ on which of those answers the combined figure of 61 for IITs, NITs and IIMs belongs to. We record both discrepancies rather than resolving them by inference.
Our reports also differ on the date of a later restatement of an independent compilation's pan-IIT total to approximately 160 — one records March 2026, the other August 2026. We state the figure as an attributed compilation claim and note that our own files disagree on the date.
Research cutoff. Research was checked on 23 and 27 August 2026. No claim was re-checked after 27 August 2026. Two reported figures were added to this page on 2 September 2026 from contemporary journalism, each labelled primary source not located.
If any claim on this page is shown to be wrong, we will correct it here, dated, with the original wording retained.
115 → ~160
a cumulative pan-IIT compilation, later restated
Restated without a published explanation of what changed, or when.
The limit of this page
Across all three institutes, not one primary document was retrieved in full for any individual death.
That is not an allegation of concealment. It is a description of a record that cannot be checked.
Who is behind this
Aarunya Issues is an initiative of Aarunya Swaraj Party, being built by students who believe that public problems deserve evidence, accountability and action.
Read the founder's noteStay connected
We are building this in public. Watch it happen.
Every issue, every representation, every response — posted as it happens. Follow Aarunya Swaraj Party and be part of the record.