AIONOS and VIKRAM SINHA’s co -founder and highest executive manager CP Gurnani on January 27, 2025, Aerocity in New Delhi, the president and highest executive officer, Indosat Ooredoo Hutchison Hutchison SIGN M. MOU with OU Sign. Photo credit: –
The rapid intake of Deepseek, a basic large -scale language model (LLM) developed in China, has AI races in a context, and the basic model developer app is also a US appistist Afronto. I am doing it.
In India, Gurgram’s based Aionos announced on Monday (January 27, 2025) that it is planning to work with DeepSeek. Outstanding Indonesia, cooperation with Telco Indosat.
Read again: What is DeepSeek and why confusing the AI sector?
The AI race faces China in the field, even in the field of restricting sanctions on the most powerful hardware, such as GPUs and processors, which are dependent on the state -of -the -art AI. I mature to the point where I was amazed at the progress.
Deepseek can demonstrate the response and analytical ability performed by Openai supported. (It is said that DeepSeek still believes that it is still hindering the lack of access to advanced chips, that the company’s CEO told Chinese officials.)
In this race, India’s status has so far as a player who has utilized downstream efficiency. As AI models become cheaper and researchers and companies can use more open source models, companies are working to use these shifts. Rather than making an expensive prepaid investment in creating a basic model.
The DeepSeek’s logo is displayed in this illustration image with the mobile phone AI assistant app | Photo Credit: Reuters
The government is focusing on creating local language solutions for translation, like Bashinii Chigi.
“Ultimately, it is necessary to reduce the cost of technology, reduce the cost of processing, and reduce the cost of expanding,” he said, the CEO and founder of Ionos, Tech Mahindra. Former CEO CP GURNANI states in briefing briefing. Monday (January 27, 2025).
When Sam Altoman visited India last year, Garnani accepted the challenge from the open CEO and built a basic model in India. It has not yet happened in the Indian industry, but Gurnani said, “Because it is a very rapid evolution industry, there will always be new developments that will occur because of new announcements every week.”
Indian companies have not yet shouted the work around DeepSeek. This can be built on it as an open source model. For example, Garnani said that while his company was cooperating with the company, the production of those efforts would be limited to Indonesia.
“Indonesia is a neutral country,” said Vikram Shinha, an Indianat CEO. Sinha, an Indian executive, led by the merger of Indonesian telephone companies, is in New Delhi in New Delhi as part of the business representative associated with President Professional President Plabowo.
Perplexity AI and inc, a major AI company, San Francisco. CEO Aravind Srinivas has indicated that these costs should not be against the construction of the basic model.
“India has to show the world that AI has an island -like foot,” said Sulinibas last week on his previous Twitter X, mentioning the recent success of the Indian Space Program. Masu.
Srinivas warned that he would fall into TRAP to avoid building basic models like DeepSeek because of his high cost.
Srinivas’s post is followed by Nandan Nilekani, a co -founder of Infosys in December. He stated that the construction of the basic model was a “not the best use” financial resources.
However, Mr. Nirkani’s point is that it is not that the money used the building more well on the existing model, but that India needs to build a computing and physical infrastructure that AI runs. did.
Disclosure -An January 28, 2025 12:25 AM IST
