In February, India will host leaders from governments, industry, academia, and civil society at the AI Impact Summit. New Delhi will join Bletchley Park, Seoul, and Paris as a venue for shaping the future of international and federal AI cooperation standards and governance mechanisms. Ahead of the event, India has published its AI Governance Guidelines, widely viewed to be the country’s blueprint for the emerging technology. Alongside these efforts and various pre-summit events, the IndiaAI Mission is pushing to create a cohesive techno-legal policy narrative, signaling that India is laying the groundwork for its AI strategy for the months and years ahead.

One of the key principles of the Guidelines is Safety, Resilience & Sustainability, drawing from the Reserve Bank of India’s FREE-AI Committee report, published in early 2025. This principle aims to “prioritise energy efficiency and frugality to enable sustainable adoption.” Surprisingly, the AI Governance Guidelines do not adequately focus on the environmental and energy impacts of AI. Beyond an indirect nod to green AI in the preface and some references to developing smaller, lightweight models in pursuit of sustainability, the guidelines stop short of outlining an actionable strategy on climate and AI. The omission adds to a list of areas where the document lacks critical depth. In a blog post, the lead author of the report acknowledges this gap, calling for future guidelines to address specific environmental risks.

Ideals must soon translate into practice. India’s climate actions have been rated ‘highly insufficient’ by the Climate Action Tracker, amid struggles to meet its climate goals. If India truly seeks to lead global conversations around responsible AI governance and exert influence at the federal and international level, it must articulate an AI climate strategy that aligns with its existing energy and climate regulatory framework, while learning from global best practices at the intersection of AI and the environment. Such efforts may help India inspire other countries in the Global South that seek to balance their twin ambitions of digital innovation and sustainable development.

Data centers risk exacerbating India’s climate crisis

Every winter, Indian and global media document the country’s worsening air quality. In 2023, the pollution contributed to about 17,000 deaths in Delhi. In November 2025, protesters rallied to demand stronger policy responses as the country’s air quality scale reportedly exceeded 500 AQI, possibly even hitting 1000 AQI. On the energy front, India has made progress towards renewables, yet fossil fuels accounted for 78% of electricity generation in 2024, according to energy think tank Ember. The International Energy Agency (IEA) states that electricity demand in India continues to increase due to industrial and domestic demand. Compounding the pressures is the environmental and energy burden posed by data centers.

In its Energy and AI report, the IEA has identified data centers as major energy consumers and noted that, while renewable generation may increase, fossil fuels will continue to play a significant role at least in the near term. In Mumbai, coal-powered data centers are affecting residents’ quality of life, as domestic and industrial demand strains the city’s power grid. Companies such as Amazon have reportedly relied on diesel generators as backup power sources. Mumbai remains attractive to global investment in data centers due to competitive energy pricing, high connectivity, and various tax exemptions, even as local communities face rising electricity bills and outages.

India’s hot climate further intensifies these challenges. Cooling requirements in high-temperature zones increase costs, strain the electricity grid further, and contribute to emissions. According to mapping by Rest of World, one-third of India’s 213 data centers are located in very hot zones.

Investigations by the media outlet Down To Earth have highlighted the severe impact of data centers on water resources in both rural and urban areas. India is home to 18% of the world's population but only 4% of its water resources, making it one of the most water-stressed countries globally. Meanwhile, water consumption by data centers in the country is expected to double from 150 billion liters in 2025 to 358 billion liters by 2030, putting further pressure on its water table. As many as 50 of India’s data centers are located in extremely water-stressed regions, and the issue is further exacerbated by a lack of transparency around water use and recycling practices. Despite these concerns, AI-related investment shows no signs of slowing, with billions of dollars flowing into the country from companies such as Google, OpenAI, and AWS for data center construction.

Across the world, similar developments are being greeted with protests over natural resources, pollution, and rising power bills. In the US, local activism has affected roughly $64 billion in data center projects, with opposition potentially influencing electoral results in several states. These instances can serve as a cautionary tale. India’s AI climate strategy must ensure that the rapid expansion of data centers does not come at the expense of water availability, or impose enormous energy and environmental costs – warnings already being raised by activists.

A spotlight on some AI and climate policies across the world

The Council of Europe’s Framework Convention on Artificial Intelligence encourages consideration of the wider risks and impacts of such technologies, including on the environment. The EU AI Act contains climate-related provisions, calling for the reporting and documentation of AI systems’ energy performance, and promoting energy-efficient general-purpose AI models. Annex XIII (c) lists estimated energy consumption as a criterion for identifying General-Purpose AI Models with systemic risks. Critics, however, argue that the Act has a regulatory blind spot on climate, prioritizing digital development over environmental concerns.

Other policy examples include the African Union’s Continental AI Strategy, which acknowledges environmental risk, and Egypt’s National Artificial Intelligence Strategy, which outlines the country’s plan on data center sustainability, green compute infrastructure, and the use of AI in the energy sector. South Korea’s Framework Act on the Development of Artificial Intelligence and Establishment of Trust considers impacting energy supply and drinking water production processes as determinants of high-impact AI.

In Brazil, Article 2 of Bill 2338/2023 sets out principles for the development, implementation, and responsible use of AI systems in accordance with, among others, environmental protection and ecologically balanced development. For example, Article 61 of the Bill calls for collaboration between the Ministry of the Environment and Climate Change and the Permanent Council for Regulatory Cooperation on Artificial Intelligence (CRIA) to promote research and certification programs to reduce the environmental impact of AI systems. However, Data Privacy Brasil, an organisation that has followed the development of the bill, highlights that text on the adoption of measures to mitigate foreseeable risks to fundamental rights, the environment, information integrity, freedom of expression, and access to information, was removed in the newer version of the Bill.

Transparency and accountability are important aspects of the net-positive AI energy framework outlined in a recent World Economic Forum report. The concept refers to the balancing act of ensuring that the energy gains enabled by AI far outmeasure the energy consumed by AI. The report highlights various tools and use cases such as public disclosure frameworks, benchmarking platforms, third-party verification, and open data repositories.

China’s Global AI Governance Action Plan calls for common standards on AI energy and water efficiency, adopting a cooperative stance on AI governance. Whether competition in the global AI Race will hamper such ambitions remains to be seen. As it stands, the false dichotomy between innovation and regulation is compromising climate goals. Case in point, the US AI Action Plan largely perceives environmental policies as an impediment to dominance and development.

India cannot afford to sideline climate considerations. Participation in the AI race should not come at the cost of an already stressed environmental ecosystem.

Path towards a cohesive AI climate strategy

India can adapt existing legal frameworks to the AI context, allowing for greater transparency for regulators and citizens. One option is to expand the coverage of the Energy Conservation Act of 2001 to bring AI data centers under its ambit, therefore requiring data centers to undergo energy audits and comply with prescribed energy consumption standards. Additionally, India can benefit from third-party auditing of AI companies and data centers on different metrics, such as energy used in the training of a model and associated carbon emissions, usage of recycled water, and how they aim to replenish or compensate the affected region, to name a few.

While using AI for climate goals is already on the docket for the Impact Summit, developing cleaner technologies and enforcing the push to renewable energy for data centers will mark a positive step in realizing India’s energy and AI ambitions. Standards should be developed for high-energy-demand AI models, informed by similar transparency practices as described above.

At present, the AI Governance Guidelines rely on voluntary measures and self-regulation, and steps must be taken to ensure that this does not lead to greenwashing measures by the private sector. This overlaps with the Indian Central Consumer Protection Guidelines for Prevention and Regulation of Greenwashing or Misleading Environmental Claims, 2024. Such interoperability with existing legislation will be pertinent as India, as of now, does not plan on introducing specific AI legislation, instead relying on the techno-legal measures already in place.

Any subsequent guidelines on AI and the environment must elaborate on concrete guidelines for the promotion of AI for Green to mitigate the harms brought about by data centers to some degree, while creating policy avenues for citizens to demand transparency from companies and governments on their energy sources and expenditure. This would align well with the Guidelines’ key principles on accountability and being understandable by design. Accountability is especially important, as recent developments threaten to weaken the country’s existing Right to Information provisions.

Furthermore, there is a need to adopt a green-by-design AI philosophy. International initiatives such as the Coalition for Sustainable AI, of which India is a member, create a common vision for environmental protection while providing a repository of evidence gathered from the private and public sectors. This is expertise India can rely on to create a cohesive AI strategy that supports capacity building, uplifting communities, and instituting legal measures for sustainable and accountable AI development.

A balancing act that India needs to get right

“Innovation over restraint” is another key principle embedded in the Guidelines. What constitutes “cautionary restraint” and what entails “responsible innovation” remains relatively vague for now, but climate considerations must not be conflated with barriers to development. The effects of a supposed AI bubble on the economy and geopolitical relations, and the risks that such technologies pose for the future, have been well-documented. Many of those risks to information integrity, education, and labor, to name a few, have received extensive coverage and are already shaping governance debates at the federal and multilateral levels. What remains missing, however, is regulatory guidance on AI and its effects on the environment, at a time when climate change is crossing alarming thresholds.

As an important player on the global stage, India has an opportunity to develop an AI climate strategy that serves as a model for other developing economies, while fostering international coordination on renewable energy and climate change. The future of AI may be uncertain, but the consequences of pollution and climate change are most certainly not. The current India AI Governance Guidelines will be further elaborated on, and when they are, the climate cost of AI must not be ignored. In pursuing technological innovation alongside climate goals, India must prioritize sustainable development, empower citizens, and protect vulnerable communities to deliver the long-lasting impacts it seeks.