CHENNAI: IIT Madras’ Centre for Responsible AI (CeRAI), part of the Wadhwani School of Data Science and Artificial Intelligence, convened a two-day Conclave on ‘Safe and Trusted AI’ on 10 and 11 December 2025, bringing together leaders from government, industry, academia, civil society and global AI institutions to advance discussions on AI safety and governance for the Global South.
The Conclave centred on the development of advanced ‘AI Safety Commons’—an open, collaborative ecosystem of shared resources—and on identifying pathways to recommend AI governance guidelines. The event aligns with CeRAI’s broader mission of fostering a global AI environment rooted in safety, reliability and trust.
The Conclave was inaugurated by Dr TRB Rajaa, Minister for Industries, Investment Promotions and Commerce, Government of Tamil Nadu, in the presence of Prof Kamakoti, Director, IIT Madras, and senior stakeholders from the Wadhwani School of Data Science and Artificial Intelligence.
Dr. Rajaa underscored the need for strong policy interventions to ensure that AI systems serve all communities, adding that Tamil Nadu’s ambition is to become an “AI-first” state. He emphasised the importance of strengthening the state’s ethical AI policy to ensure technology remains powerful and inclusive.
Prof. V. Kamakoti highlighted the growing pervasiveness of AI in daily life, noting its expanding role in crucial domains such as health and education. He stressed that safe and trusted AI has become essential as India prepares for the upcoming AI Impact Summit. The Scientific Keynote Address was delivered by Prof. Sriraam Natarajan from the University of Texas at Dallas.
Shri Mohammad Y Safirulla, Director of the IndiaAI Mission at MeitY, noted that the deliberations form part of a globally representative initiative involving experts from 112 countries. Insights from the Conclave will directly contribute to the outcomes of the Leaders’ Declaration at the India AI Impact Summit 2026.
Held against the backdrop of global momentum from major international AI forums—including the UK AI Safety Summit, AI Seoul Summit, France AI Action Summit and the Global AI Summit on Africa—the Conclave’s second day included closed-door discussions with the Safe and Trusted AI Working Group. These sessions form part of preparatory efforts for the India AI Impact Summit 2026, which will be the first global AI summit hosted in the Global South.
Speakers and participants emphasised the rising frequency of AI safety incidents worldwide and the urgent need to turn responsible AI principles into practical, context-aware mechanisms.
The Working Group, chaired by Prof. Balaraman Ravindran, aims to build global capacity, strengthen inclusion, and develop interoperable governance frameworks that embed trust, reliability and safety into AI systems.
Prof. Ravindran noted that the Working Group is committed to fostering equitable access to AI safety resources and ensuring that governance guidelines enable innovation while protecting societal well-being.
Day Two also featured a virtual keynote by Peter Mattson, Founder President of ML Commons, and panel discussions on the significance of developing AI Safety Commons for the Global South and on operationalising AI governance guidelines.
The insights generated over the two-day deliberations are expected to guide the Safe and Trusted AI Working Group’s contributions to the India AI Impact Summit 2026 and bolster India’s leadership in shaping trusted, responsible and future-ready AI systems.