With an eye toward luring more AI investment to the country, India is hosting a four-day AI Impact Summit this week that will be attended by executives from major AI labs and Big Tech, including OpenAI, Anthropic, Nvidia, Microsoft, Google, and Cloudflare, as well as heads of state.
The event, which expects 250,000 visitors, will see Alphabet CEO Sundar Pichai, OpenAI CEO Sam Altman, Anthropic CEO Dario Amodei, Reliance Chairman Mukesh Ambani, and Google DeepMind CEO Demis Hassabis in attendance.
India’s prime minister, Narendra Modi, is scheduled to deliver a speech with French President Emmanuel Macron on Thursday.
Here are all the key updates from the event:
India earmarks $1.1 billion for its state-backed venture capital fund. The fund will invest in artificial intelligence and advanced manufacturing startups across the country.
OpenAI CEO Sam Altman said India accounts for more than 100 million weekly active ChatGPT users, second only to the U.S. He also said Indians account for the most students using ChatGPT.
Blackstone has picked up a majority stake in Indian AI startup Neysa as part of a $600 million equity fundraise. Teachers’ Venture Growth, TVS Capital, 360 ONE Asset, and Nexus Venture Partners also invested. The company now plans to raise another $600 million in debt, and deploy more than 20,000 GPUs.
Bengaluru-based C2i, which is building a power solution for data centers, raised $15 million in a Series A round from Peak XV, with participation from Yali Deeptech and TDK Ventures.
HCL CEO Vineet Nayyar said Indian IT companies will focus on turning profits and not being job creators. These comments come as Indian IT stocks dip as fears of AI disrupting the IT services sector burgeon.
Vinod Khosla, founder of Khosla Ventures, said that industries like IT services and BPOs (Business Process Outsourcing) can “almost completely disappear” within five years because of AI. He told Hindustan Times that 250 million young people in India should be selling AI-based products and services to the rest of the world.
AMD is teaming up with Tata Consultancy Services (TCS) to develop rack-scale AI infrastructure based on AMD’s “Helios” platform.
Anthropic said that it is opening its first office in India in the city of Bengaluru. The company said that the country is the second biggest user of Claude after the U.S.
Anthropic is partnering with IT giant Infosys to deploy Claude models and tools like Claude code to Indian enterprises. To begin, both will deploy AI tools in the telecommunications sector with a dedicated Anthropic Center of Excellence.
Indian AI company Sarvam teases its upcoming smart glasses under the name Sarvam Kaze. The company has released several models in the past few weeks, including a dubbing model, a speech-to-text model, a text-to-speech model, and a vision model for Optical Character Recognition (OCR).
Indian conglomerate Adani said that it is allocating $100 billion to build AI data centers that would use renewable energy in India by 2035. The company said that this investment will lead to an additional $150 billion investment in areas like server manufacturing, advanced electrical infrastructure, sovereign cloud platforms, and supporting industries.
Voice AI company Cartesia is teaming up with India-based orchestrator Blue Machines to deploy voice solutions for enterprises with local data residency.
Cohere Labs launches a family of multilingual models with open weights that support over 70 languages. These models can run on local devices. The company said that it has also released models tuned to specific regions.
OpenAI said it will open two new offices in India in Bengaluru and Mumbai.
OpenAI also partnered with the Tata group to deploy 100 megawatts of compute in India with an aim to scale it up to 1 gigawatt.
India’s tech minister Ashwini Vaishnaw said that the country wants to attract over $200 billion in investment for AI infrastructure in the next two years.
India’s vibe-coding startup Emergent said that it has reached $100 in ARR and launched a mobile app.
Indian AI startup Sarvam released two new open-sourced models: Sarvam 30B and Sarvam 105B.
Sarvam also announced a partnership with Qualcomm, HMD, and Bosch to deploy its AI models to devices, including smartphones, feature phones, cars, laptops, and smartglasses.
Voice AI startup Gnani released a zero-shot voice cloning text-to-speech model called Vachana, which supports 12 languages.
BharatGen, a government-backed AI consortium, released a 17 billion parameter model called Param 2 that works across 22 languages.
Steaming service JioHotstar said it will use ChatGPT to aid content discovery with conversational search.
Sarvam launches its ChatGPT competitor called Indus that supports multiple Indian languages.
OpenAI said that 18-24-year-old users in India drive nearly 50% of the usage in India on ChatGPT.
Indian tech company Tech Mahindra released an 8 billion parameter Hindi-oriented model for educational use cases.
UAE’s G42 teamed up with U.S.-based chip maker Cerebras to deploy 8 exaflops of compute in India through a supercomputer. Abu Dhabi’s Mohamed bin Zayed University of Artificial Intelligence (MBZUAI) and India’s Centre for Development of Advanced Computing (C-DAC) are also part of the project.
On the sidelines of the India AI summit, Sam Altman said that concerns around how much water AI uses are “totally fake,” but acknowledged the issue of water usage when “we used to do evaporative cooling in data centers.”
Strangely, he also said that humans use a lot of energy as they grow up and process things around them. He thinks that the arguments around ChatGPT’s power consumption are “unfair.”
“But it also takes a lot of energy to train a human,” Altman said. “It takes like 20 years of life and all of the food you eat during that time before you get smart.”
India said that over 88 countries and organizations signed the New Delhi AI declaration for working towards using AI for social and economic good. These countries included the U.S., China, and Russia.
India joined the Pax Silica group led by the U.S. to create a smooth supply chain network of materials used in creating AI infrastructure. Other members include the U.K, United Arab Emirates, Singapore, Qatar, Japan, Israel, South Korea, and Australia.
