AI giant Anthropic's decision to take its latest AI models offline to comply with the US government's new export controls reinforces that India will have to become self-reliant in technology development, NITI Aayog member Abhay Karandikar said on Thursday.
Last week, Anthropic said it had taken its latest artificial intelligence models, Fable 5 and Mythos 5, offline to comply with a directive from the Trump administration to prevent their use by foreign nationals.
"Maybe in the development of local AI infrastructure and hardware, we are lagging...AI giant Anthropic's decision to take down latest AI models offline to comply with the US government's new export controls only reinforces that India will have to become self-reliant in technology development," Karandikar told PTI.
The export controls mark the US government's most significant step so far to restrict access to the most advanced AI models.
Last week Anthropic broadly released Fable, which is a limited version of its more advanced model, Mythos. Access to Mythos has been tightly restricted by the company due to cybersecurity fears.
A NITI Aayog member said India may have to do more to develop local AI infrastructure, hardware, because "we do not have semiconductor chips or platforms, but I think we are catching up." According to him, one of the biggest advantages that India has is its ability to deploy AI infrastructure at scale.
He further said the government is developing AI applications in some sectors, such as health, agriculture, education.
Karandikar also said the Aayog is looking for ways to deploy AI for citizen-centric services, noting that India has taken a lead in developing digital public infrastructure. He said India AI mission has supported at least 12 startups that are developing foundation models.
"The AI mission has done a lot of things in the last one year, maybe more needs to be done given the pace at which the technology is moving," Karandikar said.
Except for the headline, this story has not been edited by The Telegraph Online staff and has been published from a syndicated feed.