Author: Vedansh Gautam – Institute of Law Nirma University

In this 21st century, the dominance of Artificial Intelligence (AI) is becoming the new era for human civilization- one where machines know, act, and execute themselves at an augmented autonomy. From algorithmic recommendations on streaming platforms to autonomous drones and generative AI models like ChatGPT, AI has rapidly transformed how we work,
communicate, and govern. As the digital change creates new social contract in the world, India is at the threshold; is the ability to make use of the potentially huge capability of AI while safeguarding the three cardinal virtues for any republic: justice, liberty, and dignity: Then for me the answer remains clearly no in terms of the kind. One may build a suitable strong, yet forward thinking legal approach able of the governance of machine intelligence.

The Need for Regulation: A Digital Pandora’s Box

AI is no longer science fiction. It supports chatbots on government portals, enables judges to browse through cases and also doctors through diseases diagnosis as well for law enforcement to be used as tools in facial recognition, and predicting patrols. As well, with an increasing rate of use of its pervasive, then AI brings to us dozens of ethically and legally confusing challenges
too. If a self-driving vehicle has an accident, who will it fall on? Might the police face privacy infringement from applying facial recognition system or can facial recognition in an AI system implemented by the authorities lead a court to order an algorithm’s source code to be revealed: is biased, a decision-maker with an element of discretion should be able to take its own
decisions. This highlights a crucial notion: AI can’t be ahead of legal precedent. The unchecked AI deployment may perpetuate class divides at the same time it maintains social discriminations and undermines democratic establishments. India s digital public infrastructure dreams should go forward with safeguards, and in the future, AI better be a tool of insight not invasion, exclusions & oppression.

India’s Present Legal Landscape on AI: A Patchwork Framework

The Indian government has numerous digital governance plans and works to establish India as an AI hub worldwide but it lacks a special legislative or regulatory approach for AI use. India’s legal system to AI is disjointed and passive we can label that a “piecemeal response.” India currently has a combination of patchwork regulation of law, but not a complete system, an
assortment of disconnected laws, rather than a set of clear principles and clear rules, we have an assortment of piecemeal statutory laws without any thought of AI, large lacunae remain about oversight, about enforcement about rights.

Reliance on Outdated and Sectoral Laws

The Information Technology Act 2000, India’s basic cyberspace law, also falls before the AI era in being drafted. Cybersecurity features in addition to electronic governance and intermediary liability are features, though, algorithmic bias or autonomous machine decision- making, etc and more to come; it’s pretty silent around the room on such domains. The Digital Personal Data protection, Act 2023, also only focus on data privacy and consent. The legislation which controls the data that trains them however does not extend to governing the AI systems themselves, their decision-making processes, their influence on the lives of people, or their regulatory liabilities. Other legal domains—like the Indian Copyright Act, 1957, Consumer Protection Act, 2019,
and tort law—are similarly unequipped to deal with AI’s novel challenges. For instance, IP laws remain silent on whether AI-generated works can be copyrighted and by whom, while tort law lacks a framework for assigning liability when harm is caused by autonomous machines or algorithmic decisions.

Absence of Centralized Regulation or Institutional Oversight

There is no central regulatory authority in India that would be specifically mandated to monitor AI systems and their compliance with ethical and legal standards. In the absence of a national AI regulator, high-risk facial recognition used for policing or algorithmic screening applied to public welfare schemes is typically implemented without any transparency, accountability, or
even basic safeguards against misuse. Particularly perilous in a society as variegated and stratified as India, where algorithmic prejudice can solidify the already formidable inequalities of caste, class, gender, and region. In the absence of supervision, AI frameworks can clandestinely implant and reiterate these biases on a large scale—while staying within the bounds of law and outside the sphere of social accountability.

Legal and Ethical Challenges of Artificial Intelligence

AI poses a risk of Privacy and Mass Surveillance. The deployment of facial recognition technologies including the Automated Facial Recognition System (AFRS) by the Indian law 23 enforcement has raised alarm bells about mass surveillance. Because without appropriate legal provisions such tools go against the privacy the right to privacy recognized in Justice
K.S. Puttaswamy v. Union of India (2017).

Another challenge related to AI is of Algorithmic bias and discrimination. The objectivity of AIs is only the sum of their training datasets. Because caste, class and gender discrimination can just be embedded in historical biases in the databases training set. For example an AI deployed for loan approval will reject applicant all people from underprivileged communities
because loan data itself is discriminatory. Further when an AI-driven system causes harm, who bears the responsibility—the developer, the deployer, or the machine? The absence of legal clarity on AI liability leaves victims of algorithmic errors without remedies.

Another biggest threat of Artificial Intelligence is Labour and Automation. The number of jobless who might be displaced by automated employment could be in the tens of millions, especially in sectors like BPO, manufacturing and retail. There is also growing use of algorithmic management for its gig workforce with no legal safety net or transparency about its metrics.

International Approaches to AI Regulation: Learning from the World

The same struggle exists with AI legality in India also. No less than four have such plans, though:

European Union (EU): The world’s first general piece of AI legislation was passed late last year in the 2024 EU AI Act. Based on risk profiles (e.g. unacceptable, high-risk, low-risk), these require audits to be performed, need to be monitored by humans and require reporting:
This forbids certain forms of AI, that is social scoring. United States: Absent a nation-wide AI legislation, federal agencies enforce industry sector specific rules. The Algorithmic Accountability Act would require impact assessment for automated systems.
China: China is preparing to roll out draft guidelines on the governance of deep-synthetic and recommendation algorithm systems. It is about state’s sovereignty and AI development must fit socialist values.’
Global Standards: The OECD AI Principles and UNESCO AI Ethics Recommendations recognize and respect the rights of human beings in an approach to building trust.

Toward a Comprehensive AI Legal Framework in India

India must immediately develop a complete new legal framework that covers all challenges of AI systems. It must include those mentioned above pillars:

The Laws should recognize a separate classification of dangerous (e.g. AI in shopping online) from least hazardous (e.g. AI criminal justice) AI, and impose on these the most demanding provisions. The creation of an independent National AI Regulator. Body of technical know-how able to regulate all AI use and Enforcement of compliance. Ethicising Obligatory audits
Algorithmiques. High risk algorithms need to audited by a neutral body periodically to audit their fairness, bias etc, of high risk algorithms: that impact fundamental rights. Introduce statutory provisions for strict or vicarious liability in AI-related harm, particularly in sectors like autonomous vehicles and healthcare. Update copyright laws to address AI-generated content while safeguarding the rights of human creators. And most importantly ensure all AI use complies with constitutional guarantees of equality, freedom of expression, and dignity.

Balancing Innovation and Regulation: India’s Unique Challenge

India must balance its approach with both allowing development with strong enough constraints that curb misuse. With regards to regulation, India needs to be treading a fine line. Stricter enforcement and potential stifling of new startups as well. There are, of course, arguments that any degree of regulation is a kind of backslide to the old world. Balanced approach needs to do all three, which means for example: Allowing startups to tinker with AI in the real but controlled world of sandbox before regulatory forces take action. Use a multi-stakeholder (academia, civil society and tech industry) approach that can co-create such ethical
standards for AI. Educate judges, lawyers, lawmakers, to use them tech-wonk terms to settle AI-discrimination cases. Include representatives from under-represented groups in discussions to create guidelines for technology that benefits everyone – not just those with technical expertise.

Conclusion: Law at the Speed of Code

The advent of AI is not futuristic but a part of modern India itself. Algorithms have started controlling how the government and trade as well as social lives are run and it is up to the law to save the sovereignty of the human race while preserving democratic responsibilities. A well-regular system about AI is the essence of not just legal law but constitutional law as well. The destiny of India’s tryst with artificial intelligence, the contours of her digital republic. An opaque society or a society guided by laws about liberty equality and the right to equal treatment under law in the era of AI. The choice is what we make of how quickly and prudently we create the future of tomorrow’s technology.

References:

1. The Information Technology Act, 2000

2. The Digital Personal Data protection Act, 2023

3. Indian Copyright Act, 1957

4. Consumer Protection Act, 2019

5. Justice K.S Puttaswami and anothers. V. Union of India (2018) 1 SCC 809 (India)

6. European Union Artificial Intelligence act, 2024/1689 (EU)

7. The Organization for Economic Co-operation and Development (OECD,) AI Principles, (Issued in
2019 amended 2024)

8. UNESCO, Recommendation on the Ethics of Artificial Intelligence, (Adopted in 2021)