AI And Constitutional Values - Filling Legal Lacunae In India
Author: LawSphereX
Category: Constitutional & Human Rights
Published:
Updated:
AI AND CONSTITUTIONAL VALUES: FILLING LEGAL
LACUNAE IN INDIA
Mohit Singh, 3rd Year, B.A.LL.B. (Hons.), Assam University, Silchar Dr. Sangharsh Mishra, Assistant Professor, Assam University, Silchar
ABSTRACT
Artificial Intelligence (AI) is the ability of computers and other devices to resemble or mimic intelligent human-like behaviour. It is associated with the ability to perform tasks without any human intervention. But despite its merits, particularly in a nation like India where constitutional ideals serve as the foundation of administration, AI presents serious threats to long-standing constitutional and societal ethos. This paper delves deep into how AI threatens fundamental principles of the Constitution such as the right to privacy as well as how it will affect India’s socioeconomic and technological advancement in the future. It also critically examines the gaps within the Indian legal framework, particularly the lack of specific regulations concerning data protection and digital rights.
This paper offers an in-depth examination of the legal, policy, and technological actions necessary to tackle the challenges posed by artificial intelligence in India. Moreover, the study also investigates methods to protect constitutional rights, formulate effective strategies to combat cyberterrorism, and to ensure AI systems function with transparency and fairness. Additionally, the discussion underscores the need to combat digital gap and stresses the significance of fostering secure, ethical, and inclusive AI implementation to promote sustainable growth and democratic progress in India. Keywords: Artificial Intelligence, Data Protection, Constitutional Rights, Cyberterrorism, Sustainable Growth.
Introduction In the cradles of the digital dawn, artificial intelligence (AI)1 emerged not just as a technological wonder but as a double-edged sword that could either drive India towards remarkable prosperity or erode the foundational constitutional principles that define its democratic essence. Defined as the capacity of machines to emulate human cognition such as learning, reasoning, and self-correction, it has infiltrated every aspect of life from algorithmic credit assessments in financial technology to predictive policing in cities. With the Indian AI market expected to expand upto 7 billion by 2027,2 fueled by initiatives like the National AI Strategy3, India's determination to leverage this power for socioeconomic change is evident.
However, this optimism is counterbalanced by significant risks such as AI’s opaque ‘black box’4 nature and insatiable demand for data poses threats to established constitutional values, especially the right to privacy5 enshrined under Article 216 following the landmark Puttaswamy7 judgement of 2017, which elevated privacy8 to an essential component of human dignity. Historically, India has skillfully balanced modern-day progress with constitutional mandates by integrating principles of justice9, liberty10, equality11, and fraternity12 in its constitution adopted in 1950. The Information Technology Act 200013, represented an early digital achievement by addressing cyber threats due to the expansion of internet usage. While the
1 ENCYCLOPEDIA BRITANNICA, Artificial intelligence (AI) | Definition, Examples, Types, Applications,
Companies, & Facts | Britannica (last visited Oct. 25, 2025). 2 THE ECONOMIC TIMES, India's AI market to triple to 7 billion by 2027: Report - The Economic Times (last visited Oct. 25, 2025).
3 CARNEGIE ENDOWMENT FOR INTERNATIONAL PEACE, India’s AI Strategy: Balancing Risk and
Opportunity | Carnegie Endowment for International Peace (last visited Oct. 25, 2025).
4 Gurman Narula & Sharad Khemka, BLACK BOX AI: THE CONCERN FOR TRANSPARENCY, CENTRE FOR
LEGAL THEORY, NLIU (Oct. 25, 2025, 1:10 AM), BLACK BOX AI: THE CONCERN FOR TRANSPARENCY – NLIU-CLT.
5 Kshitiz Dubey & Dr. Jyotsna Singh, The Right to Privacy in India: Evolution and Developments, 7 Int’l J. Multidiscip. Rsch. 1, 1-7 (2025), 42002.pdf.
6 INDIA CONST. art. 21.
7 Justice K.S.Puttaswamy(Retd) v. Union of India, AIR 2018 SC (SUPP) 1841.
8 supra note 5.
9 Shafaque Ammar, Concept of Justice in the Indian Constitution, 15 Int’l J. Res. Soc. Sci. 52, 53 (2025), IJRSS6April25.pdf.
10 Utkarsh Yadav, ARTICLE 21 : A COMPREHENSIVE JOURNEY OF RIGHT TO LIFE AND PERSONAL
LIBERTY, 1 Indian J. Integrated Res. L. 1, 6-7 (2021), ARTICLE 21 : A COMPREHENSIVE JOURNEY OF RIGHT TO LIFE AND PERSONAL LIBERTY.
11 Shafaque Ammar, Concept of Justice in the Indian Constitution, 15 Int’l J. Res. Soc. Sci. 52, 54 (2025), IJRSS6April25.pdf. 12 Shafaque Ammar, Concept of Justice in the Indian Constitution, 15 Int’l J. Res. Soc. Sci. 52, 54 (2025), IJRSS6April25.pdf.
13 The Information Technology Act, 2000, No. 21, Acts of Parliament, 2000 (India).
Aadhaar14 project of 2009 utilised biometrics to promote inclusive welfare, aligning with the directive principles of Articles 3915 but also igniting privacy16 concerns under Article 2117. Later, as mentioned earlier, the Puttaswamy18 ruling of 2017 recognised privacy19 as a fundamental right. In 2018, NITI Aayog’s national strategy aimed to establish India as an “AI for All”20 centre, focusing on healthcare and agriculture to promote equality21. Recently in 2024, the ‘India AI Mission’22 gained momentum with a ₹10,371. 92 crore plan aimed at making AI accessible through startups and education, while promoting unity through ethical use.
By 2025, NITI Aayog’s report ‘AI for Inclusive Societal Development’23 provided strategies for the informal sector, and the RBI’s guidelines based on ‘Framework for Responsible and Ethical Enablement of Artificial Intelligence’24 highlighted the importance of fairness in finance. As AI25 becomes more integrated into the judiciary and governance, constitutional morality requires careful monitoring to prevent algorithmic biases, ensuring that technology enhances rather than undermines democratic values. AI Threats to Privacy & Core Constitutional Doctrines Though India is rapidly adopting artificial intelligence in its framework, its vital impact on the individual’s privacy and in other spheres of his life are already becoming visible. From automated responses to AI-based surveillance, today AI26 is playing a very crucial role in our lives.
Despite its advantages in making our day-to-day life easier, it is costing us our privacy27 14 Aadhaar (Targeted Delivery of Financial and Other Subsidies, Benefits and Services) Act, 2016, No. 10, Acts of Parliament, 2016 (India).
15 INDIA CONST. art. 39.
16 supra note 5.
17 INDIA CONST. art. 21.
18 Justice K.S.Puttaswamy(Retd) v. Union of India, AIR 2018 SC (SUPP) 1841.
19 supra note 5.
20 NITI Aayog, National Strategy for Artificial Intelligence #AIforAll (June 2018), National Strategy for Artificial Intelligence.
21 supra note 11.
22 Ministry of Electronics & IT, Cabinet Approves Over Rs 10,300 Crore for IndiaAI Mission, will Empower AI Startups and Expand Compute Infrastructure Access, Press Release 2012375 (Mar. 7, 2024), Press Release:Press Information Bureau.
23 NITI Aayog, AI for Inclusive Societal Development, (Oct. 2025),
Roadmap_On_AI_for_Inclusive_Societal_Development.pdf. 24 Reserve Bank of India, Framework for Responsible and Ethical Enablement (FREE) of Artificial Intelligence (AI) in the Financial Sector – Setting up of a committee, Press Release 2024-2025/1779 (Dec. 26, 2024), https://www.rbi.org.in/Scripts/BS_PressReleaseDisplay.aspx?prid=56612.. 25 ENCYCLOPEDIA BRITANNICA, Artificial intelligence (AI) | Definition, Examples, Types, Applications, Companies, & Facts | Britannica (last visited Oct. 25, 2025).
26 Id.
27 Kshitiz Dubey & Dr. Jyotsna Singh, The Right to Privacy in India: Evolution and Developments, 7 Int’l J. Multidiscip. Rsch. 1, 1-7 (2025), 42002.pdf.
through the following means:-
1. Biometric surveillance- It includes the use of AI technologies for collecting data
related to the movement of persons from one place to another. It includes facial & retina recognition through Al algorithms and multidimensional cameras. It also includes the use of voice identification, biometric data sharing such as fingerprints etc. In cities like Delhi and Hyderabad, surveillance mechanisms such as Automated Facial Recognition System (AFRS)28 are already being deployed without robust consent mechanisms, raising constant concerns over privacy.
2. Unauthorised data collection & Profiling- AI generally uses data received through
various means to train its algorithms, which are designed to perform various functions such as tracking and analysing user behaviour, personal interests, consumer habits, emotional and psychological patterns etc. It generally collects all this vulnerable data from the user’s life, without his knowledge and consent. It creates severe complications such as change in user behaviour due to personalised content bias, over-dependency on AI algorithms, and brainwashing etc. On the other hand, profiling is basically converting raw or fresh data into predictions based on the user such as how he might act in any particular situation or circumstances. 3. AI-generated virtual world (deepfake) & Misinformation- Deepfakes are AIgenerated media containing audio-video or photographs of persons created without their consent.
It can be offensive, defamatory and can also cause threat to the legal and political system. It includes fake AI-generated statements of political leaders, impersonating voices to commit fraud, and include the use of Generative Adversarial Networks (GANs)29, to spread misinformation.
4. Algorithmic bias & Automated decision making- The algorithmic bias emerges
basically due to faulty learning by AI systems based on biased data. Since AI learns from historical data, any existing social, racial, or gender inequality can result in faulty learning of AI. It may result in the breach of right to equality under Article 1430 of the Indian Constitution. The automatic decision, on the other hand, is a system governed 28 Elonnai Hickok et al,. Facial Recognition Technology in India, Ctr. for Internet & Soc. 3, 5-9 (2021), facialrecognition-technology-in-india.pdf.
29 Jobit Varughese, What are Adversarial Networks (GANs), IBM (May 6, 2025), Jobit Varughese.
30 INDIA CONST. art. 14.
by computer codes and algorithms which aims to take decisions on the behalf of the user, without any human intervention. Despite its convenience, it poses certain security and privacy concerns such as lack of transparency in decision making, lack of accountability and extensive data collection. Generally, this automated system collects extensive data from the user to train its algorithms, which causes serious threats to privacy rights guaranteed under Article 14. Moreover, its ‘black box’31 nature of data processing raises serious concerns related to transparency because the procedure from collecting raw data to finding the final result is generally masked. In addition to infringing upon personal privacy32, AI's interaction with the core constitutional values is evident.
Though the development of AI severely affected the pillars of privacy in India, it also affected the other important domains of the constitution such as the equality33, non-discrimination, freedom of speech and expression, and due process34 The right to equality is enshrined in Article 14 of the Indian Constitution mandates equal representation and equal protection to each and every person in the eyes of the law. Evidence such as the Crime Mapping, Analytics, and Predictive System (CPAMS)35 of the Delhi police uses historically biased crime data that overrepresents minorities which as a result produces biased outcomes, and leads to arbitrary arrest of Muslim and Sikh minorities without any individualised justification.
Similarly, the Facial Recognition Technology (FRT)36 deployed in areas such as Jahangirpuri (Delhi) showcases errors upto 35% for darker-skinned individuals, entrenching indirect discrimination. The principle of free speech outlined in Article 19(1)(a)37, which is subject to reasonable limitations under Article 19(2)38, is encountering new threats from AI-driven misinformation, especially deepfakes that create fake audio-visual content to mimic public figures or to stir unrest. These technologies pose risks to electoral fairness and societal peace, as demonstrated
31 supra note 4.
32 supra note 5.
33 supra note 11.
34 A. H. Hawaldar, Evolution of Due Process in India, Bharati L. Rev. 107, 109 (2014), EVOLUTION OF DUE
PROCESS IN INDIA.
35 R. Kavidharani, Predictive Policing and Constitutional Morality: An Evaluation of AI-Based Crime
Forecasting Technologies in India, 3 LawFoyer Int’l J. Doctrinal Legal Rsch. 903, 903-912 (2025), LIJDLR_PAPER-37-Vol-III-Issue-II.pdf. 36 Brijesh Joshi, Nasiruddin Manihar, Arif Shaikh, Facial Recognition Technology in Public Spaces, 1 VIVA- Tech Int’l J. Rsch. & Innov. 1, 1-7 (2024), MCA_23.pdf.
37 INDIA CONST. art. 19.
38 Id.
by the Cambridge Analytica39 incident, which impacted over 5 lakh Indian users by targeting voters with distorted narratives. Notable cases such as the lawsuits by Bollywood stars Abhishek Bachchan and Aishwarya Rai against deepfake pornography, highlight the reputational damage that undermines public dialogue. Moreover, deepfakes also intersect with issues of non-discrimination by targeting women and minorities for harassment and reinforcing gendered or communal biases in the expressive arena. Turning to due process40 and fair trial, procedural guarantees under Article 21 demands transparency, right to be heard, and reasoned decision-making as established in Maneka Gandhi vs. Union of India (1978)41. AI’s ‘black box’42 architecture contravanes these principles by making inscrutable decisions.
In courts, AI chatbots for case summarisation or predictive sentencing risk deskilling advocates and legal professionals, eroding human reasoning. Tools like the Supreme Court Portal for Assistance in Courts Efficiency (SUPACE)43 could generate fabricated precedents, forcing courts to verify outputs and delay justice. In bail or sentencing, such systems could assume prejudicial defaults about defendants based on demographics, violating the principles of natural justice and hence degrading the Indian justice system. AI Beyond Automation: Redefining India’s Socioeconomic Future While issues such as surveillance, data misuse, and the weakening of constitutional safeguards merit serious concern, the responsible governance of AI holds the potential to transform India’s economic, social, and technological landscape in profound ways.
As the world’s most populous country and the fastest-growing major economy, with a median age of 28 and a rapidly expanding digital infrastructure, India is in a unique position to leverage AI not only for commercial benefits but also for inclusive societal change. Although forecasts are optimistic, AI could potentially contribute $500 billion to . 7 trillion to GDP by 2035, realising these advantages requires addressing challenges such as workforce transitions and ethical governance. This narrative spans economic, social, and technological dimensions, highlighting 39 Diana Baptista, Data privacy rights stronger after Cambridge Analytica scandal, REUTERS (Oct. 31, 2022, 10:17 PM), Data privacy rights stronger after Cambridge Analytica scandal | Reuters.
40 supra note 34.
41 Maneka Gandhi v. Union of India, 1978 AIR 597.
42 supra note 4.
43 Mayank Thapa, SUPACE AI Tool in the Supreme Court of India: Transforming Legal Research and Case
Management, THE LEGAL VEDA, (Nov. 16, 2:53 AM), SUPACE AI Tool in the Supreme Court of India: Transforming Legal.
a "Solve for India"44 approach that could serve as an inspiration for the Global South. The economic impact of AI in India is significant, with estimates varying based on the pace of adoption but consistently indicating exponential growth. The NITI Aayog's "AI for Viksit Bharat"45 report predicts that accelerated AI integration could add $500-$600 billion to GDP by 2035, boosting annual growth to 8% and closing the gap to an $8. 3 trillion economy from a baseline of $6. 6 trillion. This growth is driven by productivity increase of around 15-20% in core industries through automation, predictive analytics, and R&D efficiencies. Generative AI, in particular, could unlock $621 billion in productive capacity, which is nearly 20% of the GDP of 2021.
In banking and financial services, AI-driven tools for credit decision-making and fraud detection could generate $50-$55 billion, transforming "bionic banks" that improve inclusion for 190 million unbanked adults through platforms like UPI. Manufacturing could gain $85– 00 billion from AI-optimised supply chains and mass customisation, enhancing India's share in the global value chain under the National Manufacturing Mission. 46 The pharmaceuticals sector, currently dominated by generics, could license 90-110 innovative drugs by 2034, capturing $5-$8 billion through AI-reduced R&D timelines (60-80% faster). Automotive innovations, including software-assisted vehicles (SAVs), are projected to generate $26-$33 billion from domestic subscriptions and exports, with 18-20 million units by 2035.
These gains are intertwined with India's dominance in services, where exports reached $387 billion in 2024 (11% CAGR since 2005). Global Capability Centers (GCCs), numbering over 1,700 and valued at 10 billion by 2030, are shifting towards AI engineering, multilingual prompt tuning, and governance protocols, yielding a threefold return on investment per rupee spent. However, geopolitical challenges such as U. S. export controls on AI technology could impact margins, highlighting the need for sovereign compute infrastructure under the . 25 billion IndiaAI Mission47. AI may create 69 million global roles while disrupting 83 million, with India facing 35-40% exposure in clerical and gig work (23. 5 million by 2030). Optimistically, 2. 3 million AI jobs could emerge by 2027, including oversight and ethics roles, but transitions might displace 12 million workers similar to U. S.
patterns. In conclusion, AI has the potential to transform the very essence of India, integrating opportunities across all levels of society. By 44 Alka Maurya, ‘Solve for India, Solve for the World’: Strategies of India to Lead with New Age Disruptive Technologies, 8 Int'l J. Innov. Tech. Explor. Eng'g 1151, 1151-1516 (2019), I33150789S319. pdf. 45 NITI Aayog, AI for Viksit Bharat: The Opportunity for Accelerated Economic Growth, (Sept. 2025), AI-for- Viksit-Bharat-the-opportunity-for-accelerated-economic-growth. pdf.
46 ET Manufacturing Desk, AI poised to redefine manufacturing value chain, says KPMG report, THE
ECONOMIC TIMES (May 21, 2025, 11:50 AM), AI In Manufacturing Value Chain: AI poised to redefine manufacturing value chain, says KPMG report, ETManufacturing.
47 supra note 22.
combining ambition with compassion through strategic missions, frameworks, and collaborations, India not only progresses but also sets an example for global responsible innovation, transforming dreams into lasting successes. Beyond The DPDPA: Identifying The Gaps In India’s Digital Legal Landscape The Digital Personal Data Protection Act, 2023 (DPDPA)48 is the primary legislation governing data protection in India. It is designed to oversee the handling of personal data by emphasising consent, limiting the purpose of data use, and minimising data collection. This law is complemented by the Information Technology Act49 of 2000, which deals with cybercrime and the responsibilities of intermediaries.
Constitutional Provisions, specifically Article 19 (Freedom of Speech) and Article 21 (Right to Privacy) form the basis of digital rights, further supported by IT Rules 2021 which governs online content moderation. Despite these frameworks, the implementation is inconsistent, as the draft rules for DPDPA50 was released in January 2025, and is still under review, causing delays in establishing effective protections. Due to all these inconsistencies, Indian laws face huge setbacks when it comes to data protection and digital rights. Here are the shortcomings of the Indian digital laws and it also tells where the Indian laws lag behind in the 21st century digital world.
1. Delayed and Incomplete Implementation - While the Digital Personal Data
Protection Act (DPDPA)51 was enacted in 2023, its complete and full-fledged enforcement is still pending. As its rules were drafted in 2025 and still not enforced, it leaves a huge gap and legal uncertainty for both the individuals and businesses. Until it is formally notified and enforced, the personal data is at constant risk with no accountability.
2. Government Exemptions & No liabilities - The DPDPA gives vast powers to the
government and its agencies to collect personal data without consent. It gives rise to privacy concerns and mass surveillance of individuals from specific social groups such as journalists and activists. It also contradicts with the proportionality test prescribed in the Puttaswamy52 ruling of 2017, and gives the government arbitrary powers
48 The Digital Personal Data Protection Act, 2023, No. 22, Acts of Parliament, 2023 (India).
49 The Information Technology Act, 2000, No. 21, Acts of Parliament, 2000 (India).
50 supra note 48.
51 Id.
52 Justice K.S.Puttaswamy(Retd) v. Union of India, AIR 2018 SC (SUPP) 1841.
concerning personal data and digital rights.
3. Centralisation & Government Control- The Data Protection Board of India (DPB) is
responsible for enforcing DPDPA 2023, and it is controlled and monitored by the central government. It raises concerns over its impartiality, especially when dealing with state authorities and other bodies accused of privacy and data violation. In contrast, many global data protection systems are generally autonomous from state control to ensure fairness and accountability. Examples include the European Union’s General Data Protection Regulation (GDPR)53, which works independently from any executive control.
4. Absence of AI-Centric Regulations- Major Indian laws such as the DPDPA and others
lack specific regulations related to AI governance and mapping. These laws lack provisions for regulating tasks which are performed by AI, such as the automated decision making, predictive policing, targeted advertising and manipulation, facial recognition and identification etc.. India’s framework lacks proactive checks unlike the EU’s GDPR54, this makes its approach reactive rather than preventive, leaving citizens exposed to data risks and privacy concerns.
5. Gaps in Broader Digital Rights- Apart from data protection, India lags in managing
its surveillance due to reliance on outdated laws such as The Indian Telegraph Act55 of
1885 and the Information Technology Act56 of 2000, as they lag in modern-day
provisions regarding the use of AI and implementation of robust privacy mechanisms. Cybercrime57 legislation under the IT Act58 is more reactive and does not adequately address issues like deepfakes or online harassment, while the IT Rules 2021 grant government fact-checkers the authority to suppress misinformation, raising concerns over free speech, particularly affecting journalists and activists. Furthermore, although the DPDPA59 marks a crucial advancement in forming a data
53 General Data Protection Regulation, Regulation (EU) 2016/679 of 27 Apr. 2016, L 119/1 1 (2016).
54 Id.
55 The Indian Telegraph Act, 1885, No. 13, Acts of Parliament, 1885 (India).
56 supra note 49.
57 Michael Aaron Dennis, Cybercrime, ENCYCLOPEDIA BRITANNICA (Oct. 11, 2025), Cybercrime |
Definition, Statistics, & Examples | Britannica.
58 supra note 56.
59 supra note 48.
protection framework in India, it is hindered by both structural and conceptual shortcomings. The dependence on a notice-and-concent approach does not effectively address the fundamental power disparity between individuals and data handlers, while ambiguous terms such as “reasonable safeguards” diminish the clarity of compliance. Extensive governmental exemptions pose a threat to accountability, and the lack of robust mechanisms for safeguarding children’s data may unintentionally trigger surveillance. Lastly, the reliance on a singular, centralised enforcement board lacking regional presence or proactive oversight capabilities raises doubts about the effective implementation in a nation as vast as India.
Empirical Findings: Public Perception on AI and Digital Rights in India To reinforce the doctrinal analysis of India’s evolving data protection framework, a primary survey was conducted to gauge public awareness, perceptions, and concerns regarding the AI and its relationship with constitutional norms. The survey was conducted through Google Forms and received 50 valid responses from participants across different educational and professional backgrounds. The majority of respondents were students (94%) within the 18-24 age group (92%), representing India’s digitally active demography. The questionnaire combined both close-ended as well as open-ended elements exploring the themes of AI awareness, perceived threats to constitutional values, legal literacy, trust in AI systems, and preferred safeguards.
Key findings gathered through the survey are summarised below: ● Awareness and Information While a majority of participants (54%) rated themselves moderately familiar with the term “Artificial Intelligence”60, the rest (46%) rated themselves fully aware. Most reported learning about AI through sources such as social media (82%) and news outlets (36%), rather than formal education or government sources. This reflects an informal and media-driven understanding of AI. 60 ENCYCLOPEDIA BRITANNICA, Artificial intelligence (AI) | Definition, Examples, Types, Applications, Companies, & Facts | Britannica (last visited Oct. 25, 2025).
(Graph 1) (Graph 2) ● Perceived Threats to Constitutional Values A large segment of participants perceived AI as a moderate to severe threat to
fundamental constitutional principles such as the right to privacy, equality, and freedom of expression, due process, right to dignity, and democratic participation . This aligns with the concern that unregulated AI use could hamper the basic constitutional guarantees under Article 14, 19 and 21.
(Graph 3) ● Major Areas of Concern The most alarming issues as per the data were deepfakes and misinformation (74%), data breaches (66%), and AI-enabled cyberterrorism (60%). These concerns highlight the growing public sensitivity towards AI’s role in spreading misinformation and compromising with national and personal security. (Graph 4) ● Legal Awareness Out of 50 participants, only 32% of them were aware of existing or proposed Indian laws regulating AI and data protection, while a larger part was not fully aware, revealing a substantial legal literacy gap among even educated segments of the population.
(Graph 5) ● Trust and Accountability Public belief in AI systems deployed by the government agencies such as the Crime Mapping, Analytics, and Predictive System (CPAMS)61 of the Delhi Police and the Facial Recognition Technology (FRT)62 was limited. Half of the participants expressed low confidence in such systems. Moreover, 88% agreed that AI systems affecting individuals must be transparent and explainable, reinforcing the demand for algorithmic accountability and the right to explanation.
(GRAPH 6)
61 supra note 35.
62 supra note 36.
(GRAPH 7)
● Preferred Safeguards & Reforms Participants overwhelmingly supported the strengthening of the national cybersecurity infrastructure (74%), and also demanded strict criminal liabilities for AI misuse (54%). Many open-ended responses also for the autonomous oversight bodies to regulate AI misuse, ethical AI frameworks, and an alignment with the constitutional rights, without eroding the basic constitutional guarantees.
(GRAPH 8)
A Broader Outlook
These results collectively suggest a cautious optimism, the public recognised the transformative potential of AI but remains cautious about its unchecked
expansion. The survey highlights the urgent necessity for AI governance based on rights, increased public awareness, and transparent institutional frameworks to ensure that technological advancements align with India’s constitutional values. Regulating AI in India: Policy Recommendations & Strategies Though AI offers powerful tools for public good but also creates systemic risks that current laws and institutions are ill-equipped to manage. This section aims to present practical policy interventions and regulatory strategies designed to mitigate harm, enhancing institutional resilience, and steering India towards an AI ecosystem that respects rights and encourages innovation. It outlines actionable legal reforms, governance models, and accountability mechanisms necessary to ensure that AI development remains safe, transparent, and aligned with constitutional values.
Here are some of the possible policy recommendations to strengthen India’s AI ecosystem without eroding the digital rights.
1. Robust Biometric Limits & Oversight- In respect to biometric surveillance, there
should be robust oversight mechanisms consisting of a Data Protection Board, which must be independent from any kind of direct government control. This will ensure accountability and a check on the government’s end so that no arbitrary action can be taken without a proper and justified reasoning. While storing biometric data of the users, for example in Aadhaar63, it would be better to store data in fragments rather than as a complete set. It will ensure an extra level of protection in cases of data breach and cyberattacks. Moreover, encryption of user’s data is also very important and it must be done with the help of robust computing techniques, ensuring a safe and reliable data storage. Additionally, any kind of private entity or organisation must be banned to use biometric data for user identification.
They may use an ‘One Time Password (OTP)’, which is received from the government (for ex. UIDAI) to the user as well as the organisation. Here if there are total 4 digits in OTP, the 2 digits can be sent to the user and 2 to the organisation, while entering it at the organisation’s end, the biometric authentication will become successful if all the 4 digits match exactly at the government’s end. (refer diagram 1). 63 Aadhaar (Targeted Delivery of Financial and Other Subsidies, Benefits and Services) Act, 2016, No. 10, Acts of Parliament, 2016 (India).
(Diagram 1)
2. Right to be Forgotten- In reference to unauthorised data collection and profiling, each
user must have a right to delete his/her digital footprints throughout the internet, and responsible organisations should not collect user’s data without permission. The government must make mandatory laws in order to protect citizens rights against unauthorised data collection. Additionally, India lacks in specific laws related to data localisation which makes it more vulnerable. It must be incorporated within the DPDP Act.
3. AI for AI- The rapid spread of AI-generated deepfakes, propaganda videos, and
synthetic news is eroding trust in the media. To counter this, it must be made mandatory to label ‘AI-generated content’ so that the viewers have a prior knowledge about the content. Most importantly, social media platforms should be made legally required to deploy AI detection tools and quickly remove deceptive deepfakes that violate laws (e.g. defamatory or non-consensual porn).
4. Human Intervention & Monitoring- Regarding the algorithmic bias due to faulty
learning by AI systems and automated decisions, AI implementation must be strictly monitored through human oversight for “high-risk” AI. Inspired by the European Union, India should mandate that any high-stakes AI (employment screening, loan approval, law enforcement) include human decision-makers with ultimate authority. Legally, India could follow GDPR Article 22 which safeguards the rights of individuals against automated decisions and it also restricts the use of automated decisions in high risk matters with reasonable restrictions.
5. Fully Operationalise the DPDP Act- It involves moving beyond merely announcing
it and ensuring that all key responsibilities, procedural protections, and institutional frameworks are effectively activated. This involves finalising and enforcing the related rules, setting up a fully operational Data Protection Board with sufficient independence and resources, and establishing compliance systems for data fiduciaries across various sectors. It will offer clarity to businesses, provide significant data protection rights to individuals, fulfilling Act’s intended function as India’s main governance framework.
6. Dedicated Legislation for AI Regulation- By drafting a dedicated AI law such as a
“Digital India Act” would address AI-specific issues. This could forbid the most dangerous AI practices (such as listing of forbidden AI uses and classifying sensitive AI uses under “high-risk” certifications). Apart from this, immediate modernisation of upcoming IT Rules is also very important to counter day-to-day AI risks. It must incorporate due process, narrowly define unlawful AI content and create bodies for appeal. Moreover, the government should repeal outdated IT laws that are ill suited to the digital age and introduce new laws with present and future prospects. Furthermore, ensuring that India's AI ecosystem evolves in a manner that respects rights and maintains accountability requires the promotion of strong checks and balances.
Creating independent oversight entities, like an AI Ethics Board or Ombudsperson, and involving multiple stakeholders in the rulemaking process can enhance transparency, legitimacy, and public confidence in AI governance. Additionally, building the capacity of judges, regulators, and enforcement agencies is crucial for preparing institutions to tackle new technological challenges. Mitigating Cyberterrorism Risks: Strategic Framework and Policy Responses Cyberterrorism64 is generally defined as the utilisation of computer networks and digital tools by terrorists to inflict harm, spread fear, or to pose a threat to national security. In India, Section 66F of the Information Technology Act, 2000 criminalises acts “with intent to threaten the unity, integrity, security or sovereignty of India, or to strike terror in the people”65 through computer contamination or unauthorised access.
In practice, cyberterrorism can involve the 64 Hitaishi Bharti, Cyber Terrorism in India: An Emerging Threat and its Countermeasures, 11 Int’l J. Res. & Anal. Rev. 761, 762-767 (2024), Cyber Terrorism in India..
65 The Information Technology Act, 2000, § 66F, No. 21, Acts of Parliament, 2000 (India).
disruption of essential infrastructure such as power grids, banking systems, and transportation, as well as large-scale data breaches or denial-of-service attacks targeting government systems. For instance, AI powered phishing attacks can cause huge damages to users in the form of data breaches or money related frauds which no human being could replicate in terms of sophistication, personalisation, and the ability to target thousands of individuals simultaneously. As cyberterrorism advances with the incorporation of artificial intelligence, current security systems fall short in addressing the rapid pace, magnitude, and complexity of new threats. Contemporary attacks, such as AI-powered malware and deepfake-based psychological tactics, require responses that are equally sophisticated and adaptable.
Therefore, India needs a thorough set of reforms to bolster legal readiness, improve technological capabilities, and establish robust institutional frameworks. The following suggestions present a layered approach to effectively and constitutionally prevent, detect, and counter AI-driven cyberterrorism.
1. Modernise definitions in the IT Act and BNS- Update the IT Act and BNS66 to
explicitly address cyber crimes facilitated by AI. Revise sections, particularly Section 66F of the IT Act, to include AI-driven attacks on infrastructure and to define offenses involving synthetic media, such as deepfakes used to provoke violence or terrorism.
2. Expand UAPA’s scope- Broaden the UAPA's67 reach by explicitly including
cyberterrorism in the definition of terrorist acts and granting the NIA authority to probe AI-enabled terror crimes. This might involve introducing clauses that categorize the use of digital or AI-driven misinformation to intimidate or compel as a designated terrorist activity. 3. Strengthen Content Regulation Laws- The DPDP Act already identifies nonconsensual deepfakes as data breaches, yet there is a need for stricter enforcement and broader publicity of its penalties. India should also consistently revise the IT Intermediary Guidelines to ensure AI-generated content is clearly labeled and traceable. Furthermore, a specific Synthetic Media Regulation law could require provenance
66 The Bharatiya Nyaya Sanhita, 2023, No. 45, Acts of Parliament, 2023 (India).
67 The Unlawful Activities (Prevention) Act, 1967, No. 37, Acts of Parliament, 1967 (India).
watermarking and compel platforms to identify, block, or label deepfakes.
4. Penal Enhancements & Enforcement- Increase the severity of penalties and bolster
enforcement capabilities by intensifying punishments for cyberterrorism, particularly for major assaults on essential infrastructure, through elevated sentencing standards under UAPA68 and the BNS69. Empower law enforcement with more explicit authority to confiscate digital assets, access decryption with stringent safeguards, and create a regulated safe-harbor framework for ethical security researchers engaged in supervised vulnerability testing. Cyberterrorism represents a growing threat that requires a unified approach involving law, policy, and technology. India needs to update its legal system, enhance institutional capabilities, and incorporate AI-driven solutions to identify and combat advanced digital threats. Global best practices emphasize the need for accountability, transparency, and international collaboration.
Therefore, a robust strategy should balance stringent security measures with the safeguarding of constitutional rights, ensuring that national security and democratic principles progress in tandem. Conclusion A constitutionally guided approach to artificial intelligence in India must reconcile rapid
technological advancement with the imperatives of dignity, autonomy, equality and democratic accountability. The analysis reveals that AI’s most pressing challenges i. e biometric surveillance, intrusive data harvesting, opaque profiling, deepfakes, and algorithmic bias etc. are not isolated technological mishaps but systemic risks capable of reshaping the relationship between citizen and state. These developments directly implicate the fundamental rights framework articulated under Articles 14, 19 and 21, demonstrating that the governance of AI is, at its core, a constitutional project rather than a merely technical or administrative one. India’s current regulatory landscape illustrates both forward movement and significant fragmentation.
While the Digital Personal Data Protection Act, emerging policy drafts, and sectoral rules mark an evolving regulatory consciousness, they also reveal gaps, particularly concerning the automated decision-making, transparency requirements, synthetic media
68 Id.
69 supra note 66.
standards, and robust institutional oversight. Broad exemptions, limited enforcement capacity, and the absence of dedicated AI-specific safeguards create vulnerabilities that can be exploited by both state and private entities. Without clear statutory limits on biometric systems, strong checks on surveillance architecture, and enforceable rights against automated harms, constitutional protections risk being diluted in practice. The empirical results of the document highlight the normative implications. Participants show an understanding of the risks associated with AI and express a strong need for accountability, human oversight, effective grievance procedures, and severe consequences for misuse.
Public trust is identified as a crucial element in the responsible implementation of AI, and this trust can only be established when institutions ensure transparency, fairness, and significant control over personal data. There is a clear social expectation that technological advancements should be accompanied by legal awareness, institutional responsiveness, and procedural protections. Moving forward, a coherent governance architecture must integrate constitutional values into every layer of AI regulation. Mandatory risk assessments, explainability requirements for highrisk systems, provenance norms for synthetic content, and independent regulatory bodies with technical expertise are essential.
Strengthening the DPDP framework, updating cybercrime and national security laws for AI-specific threats, and fostering judicial and administrative capacity-building will further reinforce a rights-centric digital ecosystem. Ultimately, India is at a pivotal juncture, where the potential of AI to improve public services, drive economic progress, and create new opportunities is immense. However, this potential can only be realised if AI is developed within a framework that safeguards individuals from arbitrary power, rather than fostering a wider digital divide, whether it be algorithmic or institutional. Upholding constitutional values is not a barrier to innovation; rather, it is essential for fostering a technological future that is inclusive, reliable, and democratic.
References Statutes The Constitution of India, 1950. The Information Technology Act, 2000. The Digital Personal Data Protection Act, 2023. The Indian Telegraph Act, 1885. The Bharatiya Nyaya Sanhita. 2023. The Unlawful Activities (Prevention) Act, 1967. The Aadhaar (Targeted Delivery of Financial and Other Subsidies, Benefits and Services) Act, 2016. Cases Justice K. S. Puttaswamy (Retd) v. Union of India, AIR 2018 SC (SUPP) 1841. Maneka Gandhi v. Union of India, 1978 AIR 597. Books Kshitiz Dubey & Dr. Jyotsna Singh, The Right to Privacy in India: Evolution and Developments, 7 Int’l J. Multidiscip. Rsch. 1, 1-7 (2025), 42002. pdf. NITI Aayog, AI for Viksit Bharat: The Opportunity for Accelerated Economic Growth, (Sept. 2025), AI-for-Viksit-Bharat-the-opportunity-for-accelerated-economic-growth. pdf.
NITI Aayog, National Strategy for Artificial Intelligence #AIforAll (June 2018), National Strategy for Artificial Intelligence. Alka Maurya, ‘Solve for India, Solve for the World’: Strategies of India to Lead with New Age Disruptive Technologies, 8 Int'l J. Innov. Tech. Explor. Eng'g (2019), I33150789S319. pdf. CARNEGIE ENDOWMENT FOR INTERNATIONAL PEACE, India’s AI Strategy: Balancing Risk and Opportunity | Carnegie Endowment for International Peace.