Asaduddin Owaisi, chief of the All India Majlis-e-Ittehadul Muslimeen (AIMIM) and Member of Parliament, proposed an AI-driven question bank to stop examination paper leaks.
The proposal aims to address systemic vulnerabilities in the Indian examination process. By removing human intervention from the selection of test questions, the AIMIM leader said the government can eliminate the primary point of failure that leads to leaks.
Owaisi made the suggestion during a Lok Sabha debate regarding the Public Examinations (Prevention of Unfair Means) Amendment Bill, 2026, in New Delhi. He proposed that the government develop an AI-based question bank containing 100,000 questions [1]. According to Owaisi, this system should operate with no human interface to ensure maximum transparency and security.
"We need an AI‑based question bank of one lakh questions with no human interface to stop paper leaks," Owaisi said [2].
The MP used the parliamentary session to criticize the current administration's handling of educational assessments. He said that the recurring issue of leaked papers represents a systemic failure that harms students across the country.
"The government is humiliating the youth and refusing to admit its failures," Owaisi said [3].
Owaisi said that the shift to an automated, AI-managed system would curb the risk of corruption and unauthorized access to exam materials. The proposal suggests that a large pool of questions, managed by artificial intelligence, would make it mathematically and logistically difficult for individuals to leak the specific sets used in any given exam period.
“We need an AI‑based question bank of one lakh questions with no human interface to stop paper leaks.”
The proposal reflects a growing push to replace human-centric administrative processes with automated technology to combat corruption in India's high-stakes testing environment. If implemented, a zero-human-interface system would shift the security burden from personnel management to cybersecurity, potentially reducing insider threats but introducing new dependencies on algorithmic integrity.
