When questioned on Thursday about India’s potential contribution to AI advancements, Dr. Jeff Dean, Chief Scientist of Google DeepMind and Google Research, stated that the nation is well-positioned in the field.
Dean spoke alongside eminent academic researchers, developers, and startups at the tech giant’s inaugural Research@ event in Bengaluru.
The event, which brought together the top AI practitioners and researchers in India, was aimed at investigating the country’s next frontier in AI. They talked about regional solutions and offered a forum for cooperative communication to help map out a course for utilizing AI to create a better future. Dr. Dean and other Google executives attended the event in person to talk about the future of artificial intelligence and how it will affect India.
Dr. Dean remarked, “India has a long tradition of computer science students doing amazing things; it has tremendous technical talent.” “Everyone interested in computer science is focusing on machine learning and its potential applications. India has a strong advantage in this subject due to the large number of students and professionals in the field. It is significant everywhere in the world.
Important topics for the creation of language models (LLMs) for India were also covered in the forum. This entails increasing linguistic inclusiveness to guarantee that all Indians have access to information in their mother tongue, transforming healthcare using AI, increasing the effectiveness of LLM, and giving responsible AI development first priority.
Regarding Google’s initiatives to close the digital gap, Dr. Dean stated, “One of the efforts I helped encourage was our 1,000 language effort, aiming to make AI models available in all major languages.” This is important since there is a dearth of comprehensive written content in many of the tens of millions of languages spoken worldwide. This is critical because enabling AI systems to use these languages will unleash their potential for underprivileged communities in several domains.
Important academics from the best universities in India took part in the conference as well, offering their knowledge and research on the state of AI in the region.
“This has been a fantastic gathering of India’s leading experts in AI/ML in one location,” said Balaraman Ravindran, an IIT Madras computer science professor. “I hope that this event will plant the seeds for increased industry and academic research collaboration.”Research@ events, according to Preethi Jyothi, an associate professor at IIT Bombay, are particularly helpful in bringing together Google researchers and Indian academics who are working on machine learning to explore common interests. “This involves having a group conversation about the upcoming boundaries in the field,” Jyothi stated.
Dr. Dean from Google also stressed the evolving nature of AI. He went on to say that it’s critical to understand the potential hazards and rough spots that come with growing machine learning models. When compared to these models a year or two ago, “there is a lot of progress on factual accuracy,” Dr. Dean stated. “It is imperative that we persist in optimizing the advantages while proactively addressing the drawbacks.”
Alphabet, the tech behemoth, has unveiled its most sophisticated artificial intelligence model, which can process text, audio, and video data. The much awaited AI model from Google’s owner, named Gemini, can comprehend information more deeply and with higher sophistication than Google’s previous technologies, the company stated.
The generative AI ecosystem is expanding steadily in India. Artificial intelligence technology known as “generative AI” is capable of producing a wide range of content, including text, images, audio, and synthetic data. It is anticipated to upend the pharmaceutical industry, including drug research, and the whole marketing stack in addition to generating a number of “co-pilot” prospects for numerous verticals.
After concluding its first round of fundraising, the generative AI venture Krutrim just announced that it has become a unicorn, or a business valued at over $1 billion. In December 2023, Krutrim, which means “artificial” in Sanskrit, released its foundation Large Language Model (LLM). It powers generative AI applications for all Indian languages, having been trained on the greatest representation of Indian data. This model, which was developed by a group of top computer scientists with offices in Bengaluru and San Francisco, will also drive Krutrim’s conversational AI assistant, which is multilingual and can understand and speak several Indian languages.
OpenHathi, the first large-scale Hindi language model, was released by Sarvam AI, another Indian generative AI startup. In a recent Series A round headed by Lightspeed and backed by Peak XV Partners and Khosla Ventures, Sarvam raised $41 million. Sarvam is creating a “full-stack” solution that can be utilized as an enterprise-grade platform or for training unique AI models. According to the company, a full-stack strategy will hasten the adoption of GenAI in India, where businesses are aware of the technology’s promise but are having difficulty putting it to use. BharatGPT is another LLM-based option. CoRover.ai, an AI firm, and I-HUB Anubhuti’s support system, which specializes in creating data-driven cognitive computing solutions, collaborated to create the solution.
