SDG 9: Industry, Innovation & Infrastructure | SDG 16: Peace, Justice & Strong Institutions
Institutions: Ministry of Electronics & Information Technology
At a recognition event under the IndiaAI Mission, the Ministry of Electronics & IT (MeitY) announced the approval of five projects under the second Expression of Interest in the “Safe & Trusted AI” pillar. The selected proposals focus on deepfake detection, bias mitigation in generative AI, and penetration testing for AI systems. Among them: “Saakshya” (IIT Jodhpur & IIT Madras) for multi-agent deepfake detection; “AI Vishleshak” (IIT Mandi & Directorate of Forensic Services, HP) for audio-visual forgery detection; a real-time voice deepfake detection system from IIT Kharagpur; a gender bias benchmark in agri-LLMs by Digital Futures Lab & Karya; and Anvil, an AI robustness evaluation tool by Globals ITES & IIIT Dharwad. Over 400 proposals were received from academia, startups, research organisations, and civil society. The move aims to turn safe AI governance from principle to practice—embedding resilience, fairness, and auditability into India’s AI infrastructure.
The selection signals a concrete step in building a homegrown trustworthy AI ecosystem, aligning with India’s push for AI sovereignty and responsible digital innovation.
What is the Safe & Trusted AI pillar? → A component of the IndiaAI Mission under MeitY focused on funding and developing tools for trustworthy AI - covering areas like bias audits, adversarial robustness, deepfake detection, and AI evaluation frameworks.
Relevant Question for Policy Stakeholders:
Which regulatory, funding or institutional arrangements are needed to ensure these tools scale across sectors and guard against misuse in real-world deployments?
Follow the full news here: https://www.pib.gov.in/PressReleasePage.aspx?PRID=2175698