On January 21, 2025, Aravind Srinivas, CEO of Perplexity AI and ex-OpenAI engineer, took over the social media platform X. In his post, Srinivas expressed that India could be the country that provides cost-effective solutions in the domain of artificial intelligence. He cited ISRO—Indian Space Research Organization—as an example, which has provided the world with affordable space exploration.
The following day, January 22, Srinivas announced a personal investment of $1 million and 5 hours per week to support individuals who aim to revolutionize AI in India.
He further emphasized, “Consider this as a commitment that cannot be backtracked. The dedicated team has to be cracked and obsessed like the DeepSeek team and has to open source the models with MIT license.”
Read More: OpenAI announced the GPT-4b model
In his post, Arvind Srinivas highlighted the remarkable achievement of DeepSeek, which released its model r1, outperforming OpenAI on LLM benchmarks. The r1 model offers 1 million token output at a staggering price of just $2.19. For the same output, OpenAI’s o1 model costs $60.
The existing Indian tech industry is utilizing pre-built open-source LLMs to develop applications. Disagreeing with this perspective, Aravind strives to foster fundamental LLM training in India. His vision aims to promote AI model training, which could further be released as open-source solutions.
Many professionals appreciate the tech giant’s enthusiastic view on AI model training. The post gained over a million views within two days. This shows that many individuals are intrigued by this initiative to push the boundaries of artificial intelligence in India.