We are building the national foundation to measure, evaluate, and ensure that AI systems used in Indian healthcare are trustworthy, fair, and secure.
Artificial Intelligence is rapidly entering Indian hospitals and public health programs. But a crucial question remains unanswered: How do we know these systems are safe?
Current standards only measure accuracy. They fail to test for hidden biases, data leaks, and dangerous "hallucinations" that occur when AI faces real-world Indian clinical environments.
Medical AI can unintentionally memorize and leak sensitive patient data.
Systems trained on narrow data can underdiagnose marginalized populations.
AI often invents highly confident, yet completely fabricated medical advice.
We evaluate multimodal AI systems (text, imaging, speech, and video) across four critical pillars of trustworthiness before they reach the patient.
Subjecting models to rigorous stress tests to ensure they resist adversarial attacks and comply strictly with the Digital Personal Data Protection (DPDP) Act.
Auditing systems across age, gender, language, and geography to expose disparities and ensure equitable outcomes for all Indian demographic groups.
Opening the "black box." We test whether an AI's decision-making process aligns with actual clinical logic, ensuring doctors can trust and understand the output.
Ensuring systems know when to say "I don't know." We measure a model's ability to communicate doubt and defer to a human clinician when necessary.
We are establishing the scorecards, trust labels, and regulatory frameworks required to scale Medical AI safely across India's hospitals and public health infrastructure.
*ICMR will serve as the certification authority.
For academic collaborations, institutional partnerships, or general inquiries regarding the TRUSTMAPS initiative.
Associate Professor
Dept. of Electrical Engineering (Computer Technology Group)
Joint Faculty, Yardi School of Artificial Intelligence
Laboratory for Computational Social Systems (LCS2)
Indian Institute of Technology Delhi
Hauz Khas, New Delhi, Delhi 110016, India