Tejinder Sharma, a professor at Kurukshetra University, delivered the keynote address at a special interactive meeting convened for PhD supervisors at Himachal Pradesh University (HPU) in Shimla on Tuesday, urging researchers to treat artificial intelligence strictly as a complementary instrument and never as a replacement for original thought. Speaking to a gathering of senior academics guiding doctoral scholars, Sharma argued that the integrity of academic research depends on the researcher's own analytical capacity, and that outsourcing reasoning to machine-generated outputs risks diluting the very purpose of scholarly inquiry.
The session, organised at HPU, forms part of a broader institutional effort to recalibrate research practices in an era when generative AI tools can produce drafts, summarise literature, and even simulate argumentation within seconds. Sharma's intervention highlights growing anxiety across Indian universities that students and even seasoned supervisors may lean on large language models for tasks that demand human judgement, contextual understanding, and methodological rigour — competencies that no algorithm can fully replicate, he told the audience.
ALSO READ | Apple’s First Foldable iPhone Duo Set for Tonight’s Launch with Leaked Color Lineup and Pricing
Although the full proceedings of the interactive meeting were not disclosed, the framing of Sharma's address signals a clear position from a senior academic voice: AI should function as a research aide, accelerating literature reviews, formatting citations, or identifying gaps, but the intellectual spine of any thesis must remain the candidate's own. The caution comes at a time when universities across India are grappling with plagiarism concerns that have expanded to cover AI-generated content, and when regulatory frameworks for permissible AI use in doctoral work remain uneven across institutions.
Sharma's remarks carry particular weight because they were delivered not to undergraduate students, who might be expected to misuse AI under deadline pressure, but to PhD supervisors — the gatekeepers responsible for vetting theses before submission. By addressing the audience of mentors directly, the keynote sought to shift the responsibility upstream, arguing that supervisors must themselves model disciplined AI usage and instil in their research scholars the habit of verifying, questioning, and building independently on machine-suggested material rather than accepting it at face value.
