The government has invited proposals from academic institutions and research and development organisations to build indigenously developed “tools and frameworks” that promote the just and ethical development of artificial intelligence across themes such as machine unlearning, synthetic data generation, fairness tools, bias mitigating strategies and others.

One of the tools, for example, would be around machine unlearning to rectify “inaccuracies, biases, and outdated information that can inadvertently become ingrained in machine learning models”.

“By facilitating the removal of undesirable learned behaviours, machine unlearning algorithms contribute to the development of more accurate, reliable, and fair AI systems across diverse domains,” the proposal document read.

Similarly, tools and frameworks for synthetic data generation, algorithm fairness, bias mitigation, ethical AI, and governance testing, the proposal calls for data gaps, mitigating privacy concerns, and promoting equitable representation.

Government-run institutions such as the Indian Institute of Technology (IIT), the National Institute of Technology (NIT) Indian Institutes of Information Technology (IIIT), R&D and private academic organisations that have pre-existing lab infrastructure with workstations, servers and other dedicated staff will be eligible to apply.

Such institutes must, however, justify completion of the project within the next two years, and ensure it is within the overarching goal of promoting fair and ethical AI utilisation. The proposals must be submitted to the ministry of electronics and information technology by January 12.