India won’t be second to anybody in developing AI models: IT Secretary
India will not lag in its efforts to develop foundational AI models, and those developed under the IndiaAI mission are a response to people who have questioned the country’s potential in coming up with advanced AI platforms, a senior Government official said on Friday.
While launching the country’s first video foundational model by start-up Avataar.Ai, Electronics and IT Secretary S Krishnan said that the Government has supported a range of foundational models across different applications, not just language models.
“For all those people who say that model making or these kinds of efforts cannot succeed in India, this (Avataar.Ai’s video platform) is an answer that shows that it can actually be done, and it’s an answer that shows that India won’t be second to anybody in an effort of this nature,” Krishnan said.
Avataar.Ai launched Varya, an indigenous AI video model developed with support from the IndiaAI Mission to make high-quality video generation more affordable, accessible and relevant for India.
Users can upload photos and raw videos on the platform and generate an edited version by just writing details of the desired video output.
Avataar CEO and Co-Founder Sravanth Aluru said that users will be able to create a 211-second video for every Rs 100 they spend on the platform.
He claimed that Varya will cut video generation time from 50 steps to 4, which makes it 10 times more efficient than leading models.
“We are today about 27x faster in video generation and equally 27 times cheaper in video generation. The reason why we’ve done that is we believe together as a collective that if we really want to unlock the productivity curve in India, affordable AI is very important,” Aluru said. Aluru said that the firm has developed a video AI platform by using 14 billion parameters.
“We anticipate that we’ll be unlocking markets that were not served yesterday. We’ll likely create new markets for people who weren’t doing videos because they thought it was too expensive and too cumbersome to actually start doing videos. We’ll have to see and evolve the overall impact,” he said. When asked about the impact on the jobs in the video development segment, Aluru said that there is a need to upskill people, and AI tools will need to work along with human ingenuity.
“We are allowing ideas and imagination of the creator, which is where the value lies. Convert them into actual physical videos that you can see in an efficient way. We’re removing the grunt work, but preserving the ability for someone to develop a great idea that was never thought of before. I think there’s enough originality and human ingenuity that still exists,” Aluru said.
