India is well positioned for AI advancements, says Google chief scientist Jeff Dean

“In 2012, I and a few other people started to dabble with large-scale neural networks. We were interested in how these powerful models could be used for a variety of tasks including speech recognition and computer vision, also how they could be used to better understand language and language-oriented tasks,” said Dean when asked about his views on AI development and how it all started.

Dean told the audience that he was inspired to pursue neural networks since his undergraduate days in the 1990s. He said that he was excited about the potential to train large models with more computing power. According to him, it was in 2009-12 that these neural networks began showing promising results in areas like speech recognition.

On India and aptitude for AI

“I think India is quite well positioned because you have an incredibly strong culture of respecting engineering and science and computer science.” Dean said during the conversation, adding that India is a thriving culture that respects its engineers and scientists and this will drive more students to pursue AI.

He also said that during his stay, he met some incredibly talented and eager students who are excited to learn about AI. The Google executive also mentioned that software engineering curricula in universities will need to incorporate more machine learning and AI as they are currently transforming workflows. He asserted the importance of practical experience using AI tools to solve problems for developers. According to him, sharing knowledge and following the advancements will help the community upskill.

Not replacing, but complementing

Varma, on the other hand, said that India’s approach to AI focuses more on complementing people rather than replacing them. He said that the goal is to increase productivity, efficiency, and affordability to benefit India’s large population. “I think it’s slightly different from the West. We see in the West, a lot of focus is on AI to replace. AI as a replacement tool seems to be a much more dominant discussion replacing human drivers,” he said, while contrasting the approach to AI in India with the West.

Varma also vouched for democratising access to AI tools, computing, and expertise, saying that they were critical to driving widespread innovation. He mentioned that India’s digital public infrastructure’s (DPI) efforts to expand access can combine with AI to bring the next set of million Indians into the digital economy fold. He said that areas like voice, language, health, and education are priorities where AI can help address India’s diversity and inequalities.

According to Varma, to regulate AI, the focus should be on the quality of services delivered rather than narrowly on AI companies/technologies. Democratising access facilitates a more balanced policy discourse on managing societal impacts. In essence, Varma emphasised an India-centric approach to inclusive development enabled by open and responsible AI. His comments and responses highlighted the principles of accessibility, open innovation, and balanced governance.