The rise of AI is placing immense pressure on vitality methods worldwide. As India emerges as a key participant in AI, it should discover good and sustainable methods to fulfill this rising demand.
Ann Dunkin, former chief info officer on the US vitality division, spoke with AIM on the sidelines ofthe Make investments Karnataka 2025 summit, sharing fascinating insights into how India can stability AI development with sustainable vitality options.
A Multi-Faceted Power Technique
India’s AI future largely is determined by a strong and sustainable vitality framework. Dunkin believes that the precise strategy is to leverage all out there renewable sources. “Have a look at India’s belongings and the place the nation can monetise these to get clear energy on the lowest value.”
She spoke concerning the “All-of-the-above vitality coverage” that revolves round wind, geothermal, photo voltaic, clear hydrogen, and nuclear energy.
Curiously, the time period gained prominence through the 2012 US presidential election, significantly with President Barack Obama’s administration utilizing it to stability fossil gas growth with investments in renewable vitality.
Whereas the political shift within the US has these days impacted local weather insurance policies, Dunkin stays optimistic about clear vitality’s momentum. “Even when our president chooses to not, the market will drive clear vitality ahead.”
She added that, with considerable daylight and important wind vitality potential, India is well-positioned to faucet into renewable sources to assist the power-hungry AI infrastructure.
In line with the Financial Survey 2024-2025, India’s knowledge centre market is predicted to develop from $4.5 billion in 2023 to $11.6 billion by 2032 at a compound annual development price of 10.98%.
With coal producing practically 75% of India’s electrical energy, the information centre business’s reliance on fossil fuels raises sustainability challenges. Nonetheless, companies are more and more turning to renewable vitality sources corresponding to photo voltaic and wind energy.
Hiranandani Group’s knowledge centre enterprise, Yotta, plans to safe greater than 80% of its vitality from renewable sources throughout the subsequent three to 5 years. Likewise, Hyderabad-based CtrlS is concentrating on full reliance on renewable energy by 2030.
In the meantime, Reliance Group, led by billionaire Mukesh Ambani, is buying AI GPUs from NVIDIA to construct a brand new knowledge centre in Jamnagar, Gujarat. It’s touted because the world’s largest knowledge centre. With a projected capability of three gigawatts, the power will dwarf present knowledge centres, which at present function under one gigawatt.
Jamnagar, dwelling to Reliance’s oil refining and petrochemical operations, is essential to the corporate’s renewable vitality ambitions. Close by, a 5,000-acre inexperienced vitality complicated, together with photo voltaic, wind, and hydrogen vitality initiatives, is below growth.
The Knowledge Centre Dilemma
Main tech corporations like AWS and Microsoft are investing closely in Indian knowledge centres, elevating questions on their long-term sustainability.
Dunkin identified that the main target mustn’t simply be on enlargement but in addition on optimising vitality and water utilization. “Investing in knowledge centres is sweet, however how we’re going to cut back the price, vitality consumption, and water consumption is equally necessary.”
She added that AI’s vitality footprint might be lowered by improvements in {hardware} and software program. She used DeepSeek and different fashions as an example how clients will want much less compute to run them. She additionally famous that smaller language fashions, which might function straight on gadgets, are set to grow to be extra widespread.
Security First
Past vitality, Dunkin mentioned that there’s a want for AI governance in India. She pressured the significance of information safety, regulatory safeguards, and guaranteeing AI fashions are constructed on various and unbiased datasets.
“In case your mannequin is educated by a bunch of American white males, you’re going to get a mannequin that’s biased in the direction of American white males. We want coaching knowledge from all backgrounds.”
On the Paris AI Motion Summit, Prime Minister Narendra Modi voiced related considerations about biases in LLMs.
“We should construct high quality datasets free from biases, democratise know-how and create people-centric purposes. We have to deal with considerations associated to cybersecurity, disinformation, and different threats, and should make sure that know-how is rooted in lots of ecosystems for it to be efficient and helpful,” mentioned Modi.
It turns into essential for India to construct AI frameworks that mirror its various linguistic and cultural panorama.
Agentic Methods
Trying forward, Dunkin sees agentic AI as a key productiveness driver. Nonetheless, she additionally cautioned towards knowledge privateness dangers, advocating for private AI fashions that don’t share consumer knowledge externally.
Surprisingly, Dunkin doesn’t use a lot of the AI apps like ChatGPT attributable to privateness and safety considerations. “I don’t use it so much, partially as a result of I don’t have a non-public mannequin that I’ve entry to, and so I don’t need my knowledge on the market,” she mentioned.
“Your private mannequin ought to reside in your system or in your private cloud. It ought to fetch exterior knowledge with out sending your non-public knowledge out.”
This thought aligns with India’s push for digital sovereignty and better management over knowledge privateness in AI purposes.
The publish AI Progress in India Might Be a ‘Energy’ Battle appeared first on Analytics India Journal.
