In a move that could reshape how Indians engage with AI, OpenAI has announced the launch of ChatGPT Go. With a monthly price of just Rs 399, the service promises greater accessibility to cutting-edge AI features than previously possible. This announcement will undoubtedly mark a significant shift in the way AI is used in India, not only because it makes advanced features affordable for users, but also because it makes OpenAI a more formidable rival that could jeopardize the growth of domestic LLMs.
The oligopoly like structure of the LLM market not only makes it difficult for India to make its mark, but the accompanying environmental and supply side constraints will make it even tougher for domestic market to prosper and enter into the global market.
Difficult entry barriers, lack of qualified talent, possible environmental hazards and rare earth scarcity are the four main factors affecting India's AI market. The first two are urgent and immediate. A small number of powerful companies already control a large portion of the global LLM market, and many of India's top graduates are still moving to Silicon Valley.
This imbalance is emphasized by the numbers. According to a recent survey conducted by the venture capital firm Menlo Ventures, only four companies -- Anthropic, OpenAI, Google and Meta -- control nearly 89% of the enterprise LLM usage market, with the remaining 11% going to the other companies. As of May 2025, the ChatGPT ecosystem (which includes its web and app versions as well as Microsoft Copilot) serves roughly 501 million monthly users globally, according to a report by website hosting company Hostinger. With 462 million users on ChatGPT alone, it commands a dominant 74.2% share of the LLM market.
With over three times as many users as its closest competitors (Gemini, Perplexity, and Claude), who together account for roughly 23% of the user base, ChatGPT's dominance is clear. In light of this, the market is already well-established, with strong incumbents actively gaining market share. Therefore, there isn't much to celebrate for Indian startups, even though services like ChatGPT Go, which costs Rs 399 a month, or Perplexity's collaboration with Airtel may increase access for Indians.
The two other issues facing India are supply-side and environmental limitations, both of which are less urgent but equally important. These are frequently ignored, but they will probably make a strong comeback in the upcoming years and demand policy attention.
To understand why, consider that supporting domestic LLMs necessitates data centers located in India. These facilities, the majority of which are run by cloud providers, are the foundation of large-scale AI deployments. However, data centres have a significant negative impact on the environment. The UN Environment Programme estimates that 800 kg of raw materials are needed to produce a single 2-kg computer. AI-powered microchips rely on rare earth elements, which are frequently extracted using extremely damaging methods.
Additionally, data centers produce massive amounts of hazardous electronic waste. Both during construction and after they are in use, they use a lot of water for cooling. And to power their infrastructure, they draw on vast amounts of electricity, which in most countries still relies heavily on fossil fuels.
According to the International Energy Agency, one ChatGPT request uses about ten times as much electricity as a Google search. For India, these pressures are especially severe. The nation relies mostly on fossil fuels for its energy, has limited rare earth reserves, already faces water shortages, and lacks a formalised system for disposing of e-waste.
Some would contend that these issues are neither urgent nor specific to India. Nonetheless, policymakers cannot afford to ignore the long-term environmental effects of expanding domestic AI infrastructure, especially in light of the nation's continuous problems with pollution and resource management.
Thus, India's local AI development faces two challenges: first, the market is hard to break into, and second, even a breakthrough could reveal serious flaws. India's AI aspirations run the risk of collapsing at their weakest points, which are the delicate rare earth metal supply chain and the growing environmental strains of massive data infrastructure.
If things continue as they are, India runs the risk of becoming more of a consumer than a producer of AI services. Even if it escapes that fate, the nation will probably continue to rely on imports for processor chips and rare earth metals, which leaves its AI ecosystem vulnerable right away.
Even though there are some initiatives in place to close these gaps, the rate at which AI is developing and the potential wealth that foreign players stand to gain from India's reliance on outside LLMs pose a serious threat. If India wants to achieve true AI independence, specific actions will be required.
Building a solid domestic talent pool that can create competitive LLMs is the top priority. Although Sarvam AI has demonstrated the potential of a domestic model, its uptake has been limited thus far due to the competition from multinational giants. The government must encourage capital funding, provide incentives for domestic startups, and assist in the establishment of data centers in India if the goal is AI self-reliance.
Second, we need to address the issue of rare earth dependency. To reduce its dependency on foreign countries and environmental damage, India must formalize its e-waste disposal system and guarantee that rare earths that enter its borders are thoroughly recycled.
Since using fossil fuels to power AI data centers only makes the sustainability problem worse, increasing the production of renewable energy is equally urgent. Perhaps the most difficult problem to solve is the use of water to cool AI infrastructure; this will require long-term planning and creativity (e.g., immersion cooling for data centers).
India is just beginning its AI journey, and it will not be an easy one. With the right policies, the nation has the potential to shift from a passive consumer to a defining force in the AI landscape.
(Amit Kapoor is chair and Mohammad Saad is researcher at Institute for Competitiveness).
(Disclaimer: The opinions expressed in this column are that of the writer. The facts and opinions expressed here do not reflect the views of www.economictimes.com)
The oligopoly like structure of the LLM market not only makes it difficult for India to make its mark, but the accompanying environmental and supply side constraints will make it even tougher for domestic market to prosper and enter into the global market.
Difficult entry barriers, lack of qualified talent, possible environmental hazards and rare earth scarcity are the four main factors affecting India's AI market. The first two are urgent and immediate. A small number of powerful companies already control a large portion of the global LLM market, and many of India's top graduates are still moving to Silicon Valley.
This imbalance is emphasized by the numbers. According to a recent survey conducted by the venture capital firm Menlo Ventures, only four companies -- Anthropic, OpenAI, Google and Meta -- control nearly 89% of the enterprise LLM usage market, with the remaining 11% going to the other companies. As of May 2025, the ChatGPT ecosystem (which includes its web and app versions as well as Microsoft Copilot) serves roughly 501 million monthly users globally, according to a report by website hosting company Hostinger. With 462 million users on ChatGPT alone, it commands a dominant 74.2% share of the LLM market.
With over three times as many users as its closest competitors (Gemini, Perplexity, and Claude), who together account for roughly 23% of the user base, ChatGPT's dominance is clear. In light of this, the market is already well-established, with strong incumbents actively gaining market share. Therefore, there isn't much to celebrate for Indian startups, even though services like ChatGPT Go, which costs Rs 399 a month, or Perplexity's collaboration with Airtel may increase access for Indians.
The two other issues facing India are supply-side and environmental limitations, both of which are less urgent but equally important. These are frequently ignored, but they will probably make a strong comeback in the upcoming years and demand policy attention.
To understand why, consider that supporting domestic LLMs necessitates data centers located in India. These facilities, the majority of which are run by cloud providers, are the foundation of large-scale AI deployments. However, data centres have a significant negative impact on the environment. The UN Environment Programme estimates that 800 kg of raw materials are needed to produce a single 2-kg computer. AI-powered microchips rely on rare earth elements, which are frequently extracted using extremely damaging methods.
Additionally, data centers produce massive amounts of hazardous electronic waste. Both during construction and after they are in use, they use a lot of water for cooling. And to power their infrastructure, they draw on vast amounts of electricity, which in most countries still relies heavily on fossil fuels.
According to the International Energy Agency, one ChatGPT request uses about ten times as much electricity as a Google search. For India, these pressures are especially severe. The nation relies mostly on fossil fuels for its energy, has limited rare earth reserves, already faces water shortages, and lacks a formalised system for disposing of e-waste.
Some would contend that these issues are neither urgent nor specific to India. Nonetheless, policymakers cannot afford to ignore the long-term environmental effects of expanding domestic AI infrastructure, especially in light of the nation's continuous problems with pollution and resource management.
Thus, India's local AI development faces two challenges: first, the market is hard to break into, and second, even a breakthrough could reveal serious flaws. India's AI aspirations run the risk of collapsing at their weakest points, which are the delicate rare earth metal supply chain and the growing environmental strains of massive data infrastructure.
If things continue as they are, India runs the risk of becoming more of a consumer than a producer of AI services. Even if it escapes that fate, the nation will probably continue to rely on imports for processor chips and rare earth metals, which leaves its AI ecosystem vulnerable right away.
Even though there are some initiatives in place to close these gaps, the rate at which AI is developing and the potential wealth that foreign players stand to gain from India's reliance on outside LLMs pose a serious threat. If India wants to achieve true AI independence, specific actions will be required.
Building a solid domestic talent pool that can create competitive LLMs is the top priority. Although Sarvam AI has demonstrated the potential of a domestic model, its uptake has been limited thus far due to the competition from multinational giants. The government must encourage capital funding, provide incentives for domestic startups, and assist in the establishment of data centers in India if the goal is AI self-reliance.
Second, we need to address the issue of rare earth dependency. To reduce its dependency on foreign countries and environmental damage, India must formalize its e-waste disposal system and guarantee that rare earths that enter its borders are thoroughly recycled.
Since using fossil fuels to power AI data centers only makes the sustainability problem worse, increasing the production of renewable energy is equally urgent. Perhaps the most difficult problem to solve is the use of water to cool AI infrastructure; this will require long-term planning and creativity (e.g., immersion cooling for data centers).
India is just beginning its AI journey, and it will not be an easy one. With the right policies, the nation has the potential to shift from a passive consumer to a defining force in the AI landscape.
(Amit Kapoor is chair and Mohammad Saad is researcher at Institute for Competitiveness).
(Disclaimer: The opinions expressed in this column are that of the writer. The facts and opinions expressed here do not reflect the views of www.economictimes.com)
You may also like
Gardening experts share 5 'spectacular' shrubs to add to your garden in autumn
Kaleshwaram report: HC refuses to pass interim orders on KCR's petition
Peak kisses its viral success goodbye in cheeky message before Hollow Knight Silksong arrives
'Sweat more in peace, bleed less in war': IAF Chief AP Singh
Development of trans-Yamuna area is top priority: Delhi CM Rekha Gupta
