The National Institute of Electronics and Information Technology (NIELIT), under India’s Ministry of Electronics and Information Technology, has launched Cyber Kushti 2026, a national cybersecurity and artificial intelligence (AI) hackathon designed to test participants’ judgement and decision-making skills in security assessment. According to the official announcement, the competition shifts the focus from simply identifying vulnerabilities to evaluating how participants prioritise, defend and validate security findings in increasingly AI-assisted environments.
The initiative was launched on 15 August 2026 by Prof Dr M M Tripathi, Director General of NIELIT and Vice Chancellor of NIELIT Deemed University. The hackathon forms part of the technical programme alongside the third National Conference on Cyber Security, Digital Forensics and Intelligence (NCCDFI 2026). It is being organised in collaboration with the Information Sharing and Analysis Center Foundation under the National Security Database initiative, with CERT-In acting as knowledge partner.
Competition focuses on cybersecurity judgement
Unlike conventional cybersecurity competitions that prioritise rapid vulnerability detection or capture-the-flag exercises, Cyber Kushti 2026 has been designed to assess analytical reasoning. Each participating team will receive the same intentionally incomplete “security assessment package”, containing AI-generated findings, static analysis results, architecture documentation, source code and application logs.
The organisers said the material will include inaccurate, duplicated and legitimate findings, while some real vulnerabilities will be deliberately omitted. Teams will be expected to identify, prioritise and justify their conclusions, including explaining why certain findings should be rejected.
“The most difficult part of security work has now changed. For a generation, the hardest task was discovering the problem. Today, machines generate findings much faster than any team can read them. But they still cannot confidently determine which findings are real, which are most critical and what should be fixed first. That judgement capability is what our institutions must develop, and this competition is designed to test exactly that.” Prof Dr M M Tripathi, Director General, NIELIT
The competition structure also aims to reduce reliance on expensive commercial tools by ensuring all teams receive identical datasets. Participants will be free to use AI tools of their choice during the assessment process. The launch comes amid broader national efforts to strengthen digital capabilities and cybersecurity resilience, including recent initiatives on state cybersecurity frameworks and AI capacity-building for government.
Three-stage competition format
Registration for Cyber Kushti 2026 opened on 15 August and will remain available until 10 September 2026. Participation is free for teams of up to three members, including students and independent researchers.
The competition will be conducted in three stages:
- The Akhada: An online qualifying round on 15 September 2026, from which 50 teams will progress.
- The Dangal: A second online round on 24 September 2026, narrowing the field to 10 finalist teams.
- The Kesari: The in-person final on 9 October 2026 at Dr Ambedkar International Centre in New Delhi, where finalists will present and defend their assessments before a live audience during NCCDFI 2026.
During the final stage, automated machine analysis of the same assessment materials will also be displayed, allowing delegates to compare AI-generated findings with expert human evaluation.
Recognition and controlled testing environment
Finalists will be evaluated by a joint jury comprising government, industry and academic cybersecurity specialists, including recognised scholars associated with the National Security Database programme. The winning team will receive the title “Cyber Kesari 2026”, while runner-up teams will receive the “Rustam” and “Pahalwan” honours.
Participants reaching beyond the first round will receive National Security Database recognition credits, while all entrants in the opening stage will receive participation certificates and individual scorecards. Organisers have also introduced a special award for the “most educational wrong answer”, recognising submissions that demonstrate clear reasoning despite incorrect conclusions.
NIELIT said all activities will take place in a controlled environment provided by organisers. Participants will not be required, nor permitted, to access live systems, third-party platforms or public infrastructure during any stage of the event.
This article is created with the assistance of OpenGov AI.
