NEW DELHI – Central government’s artificial intelligence initiative, IndiaAI, has selected five projects to strengthen its “Safe and Trusted AI” framework, focusing on deepfake detection, bias mitigation and generative AI security, said a statement.
The announcement follows the second round of expressions of interest (EoI) launched in December 2024 under the IndiaAI Mission’s Safe and Trusted AI pillar, which aims to ensure the ethical and transparent deployment of AI systems in India.
According to the Ministry of Electronics and Information Technology (MeitY), more than 400 proposals were received from academic institutions, startups, research organisations and civil society. A multi-stakeholder committee reviewed the submissions and shortlisted five projects.
Five projects selected for safe and trusted AI development
The selected proposals include ‘Saakshya’, a multi-agent framework for deepfake detection and governance developed by IIT Jodhpur and IIT Madras, and ‘AI Vishleshak’, a tool by IIT Mandi and the Directorate of Forensic Services, Himachal Pradesh, which enhances audio-visual deepfake and signature forgery detection using explainable AI.
Another project from IIT Kharagpur will develop a real-time voice deepfake detection system, while Digital Futures Lab and Karya will focus on addressing gender bias in agricultural AI datasets.
A fifth initiative, led by Globals ITES Pvt Ltd and IIIT Dharwad, will create ‘Anvil’, a penetration testing and evaluation tool for generative AI systems.
AI governance ecosystem
IndiaAI said the selected projects would reinforce its commitment to building a secure, transparent and inclusive AI ecosystem. The initiatives combine real-time forensics, resilience testing, and bias audits to ensure AI models used in India are accountable and reliable.
“These projects represent India’s drive to not only adopt AI but to shape it responsibly,” a senior official involved in the IndiaAI Mission said. “By developing indigenous tools for detection, testing and fairness evaluation, India aims to lead globally in safe AI innovation.”
IndiaAI’s broader mission
IndiaAI, an independent business division under MeitY, serves as the implementation agency for the national AI strategy. It works to improve access to artificial intelligence, promote ethical AI practices and foster technological self-reliance.
The Safe and Trusted AI pillar is part of IndiaAI’s multi-dimensional approach that includes building foundational models, advancing AI compute infrastructure, and promoting responsible AI adoption across sectors such as agriculture, healthcare and public safety.
In recent months, IndiaAI has also partnered with leading research institutes and private sector players to advance AI governance and data ethics frameworks, signalling a growing emphasis on balancing innovation with responsibility in the country’s rapidly expanding digital ecosystem.
The IndiaAI Mission has been approved with a budget outlay of ₹10,371.92 crore over five years.
List of selected projects for IndiaAI Mission
| Categories | Projects | Organisations |
|---|---|---|
| Deepfake Detection Tool | Saakshya: Multi-Agent, RAG-Enhanced Framework for Deepfake Detection and Governance | IIT Jodhpur (CI) & IIT Madras |
| Deepfake Detection Tool | AI Vishleshak: Improving Audio-Visual Deepfake Detection and Handwritten Signature Forgery Detection with Adversarial Robustness, Explainability & Domain Generalization | IIT Mandi & Directorate of Forensic Services, Himachal Pradesh |
| Deepfake Detection Tool | Real-Time Voice Deepfake Detection System | IIT Kharagpur |
| Bias Mitigation | Evaluating Gender Bias in Agriculture LLMs – Creating Digital Public Goods (DPG) for Benchmarking and Fair Data Work | Digital Futures Lab & Karya |
| Penetration Testing & Evaluation | Anvil: Penetration Testing & Evaluation Tool for LLM and Generative AI | Globals ITES Pvt Ltd & IIIT Dharwad |
