Can India REALLY compete in the AI wars?
Last year in June, venture capitalist Rajan Anandan asked OpenAI’s CEO Sam Altman a loaded question: Can India build an artificial intelligence tool like ChatGPT? Altman’s responded that he finds it a little hopeless! Satyam’s former CEO, CP Gurnani, wasn’t about to take this lying down. He tweeted that he accepts the challenge.
Altman tried to explain, by saying that he was talking about India’s funding capacity, and not the potential of Indian AI enthusiasts. But if you think about it critically, you might realize that Sam was kind of right. Before we dive into the numbers, let’s rewind and understand how we exactly is we got here. How did AI Involve into this tech juggernaut?
Importance of Research
The current AI hype, powered by these Large Language Models or LLMs, started with a research paper. It came from the University of Toronto, authored by Geoffrey Hinton- the “Godfather of modern AI” – along with Ilya Sutskever of Open AI fame and Alex Krizhevsky (that’s where the name AlexNet comes from). They took an old AI algorithm and sort of supercharged it using just two NVIDIA GPUs.
But here’s the key word: RESEARCH. How many Universities do you know in India known for their cutting edge, competition winning or patent awarded, commercially implementable research in Artificial Intelligence?
India’s Resource Constraints
India does not have the resources; because India ALWAYS exports it’s best resources.
look at the top AI researchers around the world almost everyone belongs to the Silicon Valley. And if you go through the history, you will understand that the researchers or the scientists are the main players of the entirety of the AI ecosystem. Frank Rosenblatt, who invented the perceptron, Geffory Hinton who gave the idea of back propagation algorithm, Yann LeCun who developed the Convolutional Neural Network or Fie-Fie-Li who created images dataset platform ImageNet – they were all researchers. Now if you see, these people belong to the USA and the universities out there. Although Geffory is from UK and LeCun from France, later they shifted to the US. Let’s call it the Silicon Valley Talent Vacuum.
The Silicon Valley Talent Vacuum
This vacuum isn’t just sucking in researchers. Look at the CEOs of major tech companies: Sundar Pichai at Google, Satya Nadella at Microsoft, Shantanu Narayen at Adobe. What do they have in common? They’re all of Indian origin. But that’s the word origin. This CEOs are also part of a four million-strong minority group that is among the wealthiest and most educated in the US. About a million of them are scientists and engineers.
In fact, Indians make up a huge chunk of the tech workforce in the US. over 70% of H-1B visas – you know which were work permits for skilled foreigners – go to software engineers. And in tech hubs like Seattle? Almost 40% of foreign-born engineers are from India.
So why is this happening? Two words: money and opportunity.
Between 2013 and 2023, the US poured a mind-boggling $335 billion into AI. And India? Just about $9 billion. That’s less than 3% of what the US invested. It’s like bringing a slingshot to a laser gun fight. And the gap is only widening. In 2023 alone, US AI startups raised $16.2 billion Dollars. In India $113.4 million. That’s less than 1% of what the US raai NEWSised in a single year.
This brings us to our second factor: the infrastructure
The backbone of modern AI isn’t just brainpower – it’s computing power. And who controls that? Well largely the US tech giants. Amazon Web Services, Google Cloud, Microsoft Azure – these American companies dominate the cloud computing market, providing the raw power needed for AI development. But there’s a twist in this tale. These cloud giants aren’t just selling services – they’re investing in AI startups, who then become their customers. It’s a cycle that some call “revenue round-tripping.” But that the different topic all togheter.
Take Amazon’s $4 billion Dollar investment in Anthropic, an AI startup. Part of the deal? Anthropic uses Amazon’s cloud services. Google and Microsoft have similar arrangements with other AI companies. Oracle also became a cloud partner for Cohere after investing in the startup. This creates a closed loop that’s hard for outsiders to break into. And it’s not just about the software. Nvidia, the company that makes these powerful chips. You know what their market cap is? It’s almost as much as India’s entire GDP. One company, less than 30,000 employees, is worth nearly as much as a country of 1.4 billion people.
So where does this leave India?
India’s AI Potential
India gives birth to the talent. But that talent is being sucked away by the Silicon Valley Vacuum. India lacks the massive investments and infrastructure needed to compete at the highest levels of AI development. But here’s where the story takes another interesting turn.
Remember how we talked about the early days of SaaS in India? Well, the AI startup scene in India today is reminiscent of these times. Funding might be constrained, but there’s a buzz of activity. Take a look at these startups: Krutrim, India’s first AI unicorn, is valued at $1 billion. Sarvam AI, building language models for Indian languages. Mad Street Den, is creating AI solutions for global enterprises. These startups are doing something interesting. They’re not just trying to replicate what Silicon Valley is doing. They’re solving uniquely Indian problems, like creating AI models that understand India’s very many languages. Dev Khare, a partner at Lightspeed Venture Partners India, puts it this way:
You have to take a risk and say, ‘This is where the world will be in a few years from now. That market doesn’t exist today, but I’m goanna bet it exists. I’m going to build for that.’ That’s a bit of a newer DNA for India.”ai
So, can India compete in the AI wars? The odds are stacked against it. The funding gap is massive, the brain drain is real, and the infrastructure lag is significant. But just when you think India story’s over, there’s a plot twist.
Remember CP Gurnani, the CEO who challenged Sam Altman’s claim? Well, he wasn’t just talking big. He was cooking up something in the background. In just five months, and with less than $5 million – that’s pocket change in Silicon Valley terms – Tech Mahindra developed an Indian Large Language Model. And not just for one language, but for local languages and over 37 dialects.
This project, called Project Indus, isn’t just about proving Altman wrong. It’s about something bigger. It’s about being “Atmanirbhar” or self-reliant. They’re teaming up with big names like Dell Technologies and Intel to make it happen. The goal? To bridge the linguistic gap and let millions of Indians interact with AI in their native tongues. This is where India might have an edge. While Silicon Valley is busy creating one-size-fits-all solutions, India is focus on its unique challenges. It’s not just about competing – it’s about solving problems that matter to over a billion people.
India’s got something unique. It’s got a massive pool of engineering talent, a huge domestic market to test new ideas, and now, a growing ecosystem of AI startups that are taking on challenges that Silicon Valley might overlook entirely As Gurnani puts it, the secret sauce for Indian companies is
“frugal plus innovation plus technology plus people leadership.”
It’s how companies like IndiGo and Airtel compete with giants. And now, they’re bringing that same spirit to AI. The AI race isn’t just about who has the most money or the biggest computers. It’s also about who can solve real problems for real people. And in that race, India might just have a fighting chance.
The question is: will it be enough to make a real impact on the world stage? Only time will tell. If you want to know more about how the global AI landscape is shaping up, click on this link.
