New Delhi, February 20 Institutions should not be overly enthusiastic about using AI-based solutions until they have been tested and proven reliable, a government official said on Friday.
Rohit Bhardwaj, Deputy Director General, Data Informatics & Innovation Division, Ministry of Statistics and Programme Implementation (MoSPI), also suggested that government departments should make relevant data AI-ready by creating files that can be read by a computer.
"There should be context files, there should be semantics, and there should be metadata," he said, emphasizing the importance of storing information in a structured format so that it becomes AI-ready.
Bhardwaj highlighted the importance of testing an AI solution before putting it into use, citing a report by a group of researchers from a Canadian university, which showed that AI can analyze a particular dataset in multiple ways even when given similar prompts.
"I just want to emphasize that we should not be overly enthusiastic about things that are still untested," he said during a session on 'AI-Ready Data: Shared Infrastructure for Innovation' at the AI Impact Summit.
"I would be the first person to adopt AI for my work, but it needs to be reliable," he added, "People are not aware of what it takes to make data AI-ready, and it is the responsibility of institutions like MoSPI to make people aware of what AI readiness is all about."
He suggested that ministries or government departments should have a catalog, and that all files should not be in PDF format, as they should be machine-readable.
"I plan to create a slide deck, showing what AI can see and what AI cannot see. So, if my folder has 10 versions, some answers will come from Version 1 and some from Version 2. Unlike humans, AI is designed to scan the entire available data," he said.
Prem Ramaswami from Google said that his company tries to bring multiple datasets globally together in a common knowledge graph and put a search engine on top of that. "So that you can quickly access that data. We make this entire stack open-source."
He pointed out that the idea of centralizing data at one source is dangerous.
Instead, he suggested, the data should be located at every organization and governed locally by the organizations, making them affordable and accessible to business enterprises.
"If you have 74 million MSMEs in India, you cannot afford to hire data scientists or computer scientists," he said.
He further explained, "If you are a policymaker, thinking about poverty, climate change, education, and health, these are holistic problems. It is no longer... I can go to one ministry, pull one spreadsheet, and solve the problem of poverty. We should approach AI as a tool that we can use to derive the answer," he said.
