IndiaAI Scales Up Safe AI Efforts with Deepfake Detection, Bias Mitigation, Generative AI Security Tools
India’s flagship artificial intelligence initiative, IndiaAI, has selected five groundbreaking projects to advance its Safe & Trusted AI mission.
New Delhi, October 8: India’s flagship artificial intelligence initiative, IndiaAI, has selected five groundbreaking projects to advance its Safe & Trusted AI mission—marking a significant step toward strengthening the nation’s AI security, transparency, and governance ecosystem.
The projects—spanning deepfake detection, bias mitigation, and AI penetration testing—were chosen from over 400 proposals submitted by top academic institutions, startups, and research organisations under the second round of the Expression of Interest (EoI) launched on December 10, 2024.
The selection was made by a multi-stakeholder expert committee that evaluated proposals on their technical innovation, societal impact, and scalability. Together, the chosen initiatives aim to translate the vision of Safe & Trusted AI into practice through resilience testing, bias auditing, and forensic AI capabilities.
Five Projects to Strengthen India’s Safe AI Framework
- Saakshya – Developed jointly by IIT Jodhpur and IIT Madras, this multi-agent, retrieval-augmented (RAG) framework is designed for deepfake detection and governance. It will provide a national-level toolkit for identifying synthetic media and combating AI-driven misinformation.
- AI Vishleshak – A collaboration between IIT Mandi and the Directorate of Forensic Services, Himachal Pradesh, this project enhances detection of audio-visual deepfakes and signature forgeries using adversarial robustness and explainable AI to ensure reliability in forensic investigations.
- Real-Time Voice Deepfake Detection System – Created by IIT Kharagpur, this system uses advanced signal processing and machine learning to flag voice-cloned content in real time, bolstering the country’s AI forensics infrastructure.
- Bias Mitigation in Agriculture AI – A joint initiative by Digital Futures Lab and Karya, this project evaluates gender bias in agricultural large language models (LLMs) and develops Digital Public Goods (DPGs) for fair data benchmarking and inclusive AI deployment.
- Anvil – A partnership between Globals ITES Pvt Ltd and IIIT Dharwad, this penetration testing and evaluation tool will probe generative AI models for vulnerabilities, strengthening resilience against misuse and adversarial attacks.
Towards a Safe and Inclusive AI Ecosystem
IndiaAI said the selected projects represent the country’s growing capability to build indigenous, secure, and socially aligned AI solutions. “These initiatives will fortify India’s AI landscape with robust deepfake detection, bias mitigation, and responsible testing frameworks,” the agency stated.
The mission aims to ensure that AI systems deployed across governance, business, and society are ethical, explainable, and resilient, aligning with India’s commitment to AI safety, accountability, and inclusivity.
About IndiaAI
IndiaAI, an Independent Business Division under the Ministry of Electronics and Information Technology (MeitY), serves as the implementation agency for the IndiaAI Mission. Its mandate is to democratize AI’s benefits, promote technological self-reliance, and ensure the ethical and responsible use of artificial intelligence to serve the diverse needs of the nation.
