Author: Mehak Verma
Indian Institute of Management Rohtak
To the Point
We all have seen how rapid the growth of Artificial Intelligence (AI) has been, whether it is in the HR department or logistics, it is being used everywhere. However, if we have to gain some, we have to lose some, and while this growth of AI infrastructure, including energy-intensive data centers, hardware production, and an increase in demand for electricity and electronic waste, raises urgent environmental concerns, we have lost nature while being technologically advanced. The environmental costs of technological expansion often go unnoticed, falling into a governing blind spot. This article examines how existing environmental laws address the ecological impact of AI development and deployment. It oversees how the challenges in applying traditional environmental frameworks to AI infrastructure and calls for a reframing of regulatory provisions to ensure a sustainable technological future.
Abstract
As AI technologies continue to expand, their environmental footprint grows proportionally. The ecological implications of AI, ranging from energy and electricity consumption to toxic e-waste, pose significant challenges to environmental sustainability. This article explores the interface between environmental law and AI development, with a focus on applying existing legal doctrines such as the Polluter Pays Principle and Environmental Impact Assessments to the AI industry, as there is no provision under the Environment Law for the damage caused by Artificial Intelligence as of now. This article also highlights the urgent need for reform in environmental governance to accommodate the scale and speed of technological progress while maintaining sustainability.
Use of Legal Jargon
Now, all the jargon used further is not implemented in AI-related technologies, as people are still trying to figure out how to adjust AI to ensure sustainability and apply environmental laws. But even though it is not explicitly mentioned about AI on the side of giving damages, there is still an environmental impact caused, and the whole point is that laws are needed to protect the environment, either way. So we can see some jargon relating to AI and sustainability,
Polluter Pays Principle (PPP): Environmental costs like emissions and e-waste must be borne by those causing them, including AI-based enterprises.
Precautionary Principle: Regulatory intervention is justified even if environmental risks of AI infrastructure (e.g., data centre water use) are not fully proven scientifically. Prevention is better than a cure.
Environmental Impact Assessment (EIA): Mandatory for large-scale infrastructure like hyperscale data centres; evaluates impacts on air, water, biodiversity, and local communities.
The Proof
Large-scale AI arrangements mostly run in data centres, which have a significant environmental impact. Data centres generate electronic waste, or e-waste, which can contain dangerous substances like mercury and lead.
The data centres use large amounts of water both during construction and to cool down the systems while its running. One estimate suggests that globally, AI infrastructure may soon use six times more water than Denmark, a country with 6 million people. This is especially concerning since one in four people worldwide still lacks access to clean water and sanitation.
Powering AI technology takes a lot of energy, and a lot of that energy still comes from fossil fuels, which release greenhouse gases that cause climate change. The International Energy Agency (IEA) reports that a single request made through ChatGPT uses 10 times more electricity than a Google Search. In Ireland, a growing tech hub, the IEA estimates that data centres could use nearly 35% of the country’s electricity by 2026.
Due in part to the growth of AI, the number of data centres has exploded from 500,000 as of 2012 to 8 million today. Experts expect AI’s demand on natural resources to keep rising which is a cause of concern.
Case Laws
Vellore Citizens’ Welfare Forum v. Union of India (1996) 5 SCC 647
The Supreme Court addressed pollution caused by tanneries in Tamil Nadu and emphasized the importance of sustainable development. It applied the Precautionary Principle, which requires preventive action in the face of environmental harm even without full scientific certainty, and the Polluter Pays Principle, which holds polluters financially responsible for the damage they cause. The Court found the tanneries liable, ordered them to pay fines and compensate affected people, and directed the Central Government to set up an authority for environmental protection. These principles are also relevant today in regulating the AI industry, which may pose uncertain but potentially serious environmental risks.
Sterlite Industries (I) Ltd. v. Union of India (2013) 4 SCC 575
This case emphasized corporate accountability for environmental degradation and affirmed the necessity of strict compliance with EIA norms. Just like in this case, the Sterlite plant caused environmental harm despite having legal clearance, AI technologies can also cause harm even when they follow current laws. The case shows how systems meant to protect the environment were not always strong enough, and action came only after damage was done. The same risk exists with AI if we don’t have strong rules and monitoring in place, as it can harm society before the problems are fully understood. Just like the court imposed penalties and called for monitoring, we also need to create safeguards and hold AI companies accountable for the consequences of their technologies.
T.N. Godavarman Thirumulpad v. Union of India (2002) 10 SCC 606
The Court invoked the Public Trust Doctrine, recognizing the State’s obligation to protect environmental resources. The doctrine can be extended to assess whether government permissions for energy-guzzling AI infrastructure violate public trust.
Municipal Council, Ratlam v. Vardhichand (1980) 4 SCC 162
The Supreme Court made it clear that public authorities cannot ignore their duty to protect public health, even if they don’t have enough money. The Court said that people have a right to live in clean and healthy surroundings, and the government must act to stop public nuisance. This principle can also apply to AI-related facilities. If an AI data center or factory causes too much noise, heat, or toxic discharge, it becomes a public nuisance.
Conclusion
The current environmental law administration, although practically robust, is structurally under-equipped to handle the environmental consequences of AI proliferation. AI uses a lot of energy and creates electronic waste, harming the environment. Existing laws like Polluter Pays, Extended Producer Responsibility, and the Precautionary Principle help, but they need updates for AI-related issues. Governments should set clear rules, check AI data centers for sustainability, and make tech companies manage waste better to reduce harm. Only through such legal evolution can the promise of AI coexist with ecological preservation. Thus, the only request is to use this technology only when in need of urgency and not for unnecessary usage, because your work might get done, but there will be no world left to see the result of the work you have done using ChatGPT.
FAQS
1. How does AI impact the environment?
AI infrastructure, such as data centers and hardware production, consumes vast amounts of energy and generates e-waste. Training large AI models emits significant carbon dioxide, contributing to global warming.
2. Are there specific environmental laws for AI?
Not yet. Current environmental laws like the Environment (Protection) Act, 1986, and E-Waste Management Rules, 2016, indirectly apply to AI infrastructure but lack explicit mention or standards specifically for AI systems.
3. Can companies be held liable for environmental harm caused by AI operations?
Yes it can. Under principles such as the Polluter Pays Principle and statutory provisions, tech companies can be held accountable for emissions, waste, and other environmental externalities caused by their AI-related activities.
4. What role does the judiciary play in regulating AI and environmental sustainability?
Courts have laid down essential principles such as PPP and the Public Trust Doctrine, which can be extended to AI-related cases. Judicial intervention is crucial in bridging legal gaps until legislative reform occurs.
