Building foundational AI models demands massive investment, a challenge for Indian startups. While unable to compete with US giants like OpenAI and Anthropic in funding, Indian firms are innovating. They are focusing on creating Indic language mod...
By ETtech |
TIL Creatives
I observe the vast difference between high-end and low-budget AI training setups. The contrast in resources is evident.
Billions. That is how much it has taken to build foundational AI models for OpenAI and Anthropic. There is no end to their fund raise, given the latter’s $13 billion round or OpenAI’s $500 billion Stargate project with Softbank. Clearly, building foundation models require deep pockets, an advantage the Indian startups do not possess. As former Infosys CEO and founder, Vianai Systems Vishal Sikka pointed out, while we cannot compete with the US players, India as a society has been resourceful and needs to get creative. ET looks at how Indian firms are building Indic language models at a fraction of the cost.
