ChatGPT’s parent company OpenAI has pledged its support to the India AI Mission by participating in the Application Development Initiative (ADI) under the programme, a top company executive said on Wednesday.
“OpenAI is committed to supporting the India AI Mission’s app development initiative to ensure that Indian developers can build on our models and deliver social benefits at scale. We look forward to continuing the conversation with the ministry and assessing where we could provide the most value,” said Srinivas Narayanan, Vice President, OpenAI.
Identified as one of the seven pillars of India’s AI Mission, ADI aims to promote AI applications in critical sectors by addressing issues of central ministries, state departments and other institutions.
Speaking at the Global India AI Summit in Delhi, Narayanan said the company is keeping India in mind while taking important decisions in the area of large language models (LLM).
“Entrepreneurs understand the gaps in the market. They are creating innovative products, and tools like ChatGPT are helping them accelerate this process in ways that have never been seen before. We are reducing the cost of AI, enabling developers to write code, and helping them create fully conversational and natural computing interfaces,” he added.
Narayanan also said that cost and language were the two most important factors for the business in India.
“There are a couple of things that are specific to the Indian context: language is obviously our priority and cost. Those are the two pieces of feedback that we’ve gotten from the community over the last year. GPT-4 addresses a lot of that feedback,” he added.
He said the company has improved tokenization, due to which it has been able to use three times less tokens in the latest model, for processing Indian languages.
Tokenization is the process of breaking down a piece of information, such as a sentence or paragraph, into individual words or “tokens.”
These tokens are the building blocks of language, and tokenization helps computers understand and process human language by breaking it down into simpler units.
First published: July 3, 2024 | 6:15 p.m. IST
