AI Is Already Deciding Who Gets Hired, Credit, and Surveilled
Artificial intelligence is already deciding who gets hired, who gets credit, what content reaches you, and who gets surveilled. The question is no longer whether AI will transform corporate India, but whether our laws can keep pace. For a nation whose Supreme Court anchored equality in dignity and due process in Justice K S Puttaswamy's privacy judgment, the absence of binding AI regulation is not policy lag. It is the most urgent need of the hour.
AI Adoption in Indian Boardrooms and Its Impact on Employment
AI is no longer a pilot project in Indian boardrooms. It is the operating system. IT services firms automate code review. Banks deploy machine learning for credit decisions. Gig-economy giants route delivery riders through optimisation engines that treat humans as network nodes. NASSCOM data confirms rapid adoption, with entry-level jobs that absorbed millions of Tier-2 and Tier-3 graduates facing the sharpest automation risk. In a country adding ten million workforce entrants annually, this is an immediate restructuring of employability itself, further requiring a deep dive in curriculums in engineering colleges.
Algorithmic Harms: Invisible Violations of Article 14
The harms extend beyond displacement. In warehouses outside Delhi and Chennai, algorithms set picking targets and flag 'inefficiency' without human recourse. AI-driven credit scoring through platforms like Jio and PhonePe has expanded inclusion, but a rejected loan application comes with no explanation and no appeal. When an opaque system penalises a borrower for a postal code or a data proxy for caste, it violates Article 14 in ways invisible and unaccountable.
India's Fragmented Regulatory Architecture
India's regulatory architecture touches AI at its edges. The IT Act, 2000, the Bharatiya Nyaya Sanhita, 2023, and the DPDP (Digital Personal Data Protection) Rules, 2025 create overlapping zones of accountability. The RBI, SEBI, and health regulators have issued guidelines. The DPDP Rules mandate algorithmic fairness assessments for significant data fiduciaries. This is progress, but it is not governance.
Voluntary Guidelines vs. Binding Law: The India AI Governance Guidelines
In November 2025, MeitY (Ministry of Electronics and Information Technology) published the India AI Governance Guidelines built on seven ethical 'sutras' and made them entirely voluntary. No organisation can be fined for non-compliance. This is 'innovation over restraint,' a deliberate contrast to the EU AI Act, which carries binding risk tiers and penalties of up to 7 per cent of global revenue. A private member's AI Bill proposes penalties up to Rs 5 crore, signalling that the net is tightening, but not yet tight enough.
Structural Deficit: No Standalone AI Regulator in India
The structural deficit is clear. India has no standalone AI regulator capable of auditing complex models. No mandatory conformity assessment for high-risk systems, no prescriptive risk classification — only thematic guidance that leaves the same algorithm judged differently by different bodies. The DPDP Act governs data misuse, not automated decision-making. For a country whose digital public infrastructure serves over a billion people, integrating AI governance with DPI (Deep Packet Inspection) remains aspiration, not architecture.
AI Governance as the Next Frontier of Privacy and Dignity
AI governance is the next frontier when it comes to privacy being intrinsic to dignity. When an algorithm denies credit without explanation, it violates self-determination. When an employer monitors keystrokes and sentiment, it erodes workplace dignity. When biased hiring tools exclude marginalised candidates, they encode discrimination into infrastructure. Law promises equality; if the law cannot inspect the algorithm, the promise is hollow.
Urgency for a Young Population in Economic Transition
India's demographic trajectory makes this urgent. We are a young population in economic transition, a workforce migrating from farms to gig platforms, a polity negotiating digital caste and gender inequities. Global standards are a necessity, not a luxury. Without binding algorithmic impact assessments, we encode historical discrimination into HR tools. Without transparency mandates, 'black box' models deepen exclusion. Without a unified enforcement body, workers harmed by algorithmic management have no recourse.
From Principles to Prescription: What India Must Do
India need not copy Brussels completely. But we must move from principles to prescription. AI systems used for hiring, credit, or criminal risk should face mandatory pre-deployment impact assessments, repeated annually, with attention to caste, gender, and language bias. The proposed Digital India Act must create a dedicated AI oversight body with authority to withdraw harmful systems. Corporate transparency must be non-negotiable: every worker managed by an algorithm deserves to know how performance is measured, and every consumer denied service deserves a human appeal. Education must pivot to AI literacy so that ten million annual workforce entrants can govern AI rather than be governed by it.
Lessons from India's Technology Regulation History
The pattern of Indian technology regulation is instructive. The IT Act began as guidelines. The DPDP Act began as voluntary code. In each case, harm accumulated in the gap between adoption and enforcement. AI moves faster than any predecessor. The lag between voluntary guidance and binding law is not a prudent pause. It is an open window for structural damage to employment, equality, and dignity, which Justice Puttaswamy placed at the heart of our constitutional order.
The Risk of Becoming a Testing Ground for Risky AI
If Parliament does not act with speed, India will face algorithmic harm, public scandal, and reactive legislation drafted in crisis. The EU is exporting its regulatory model through trade agreements. The United States is moving toward federal AI legislation. China operates a state-directed framework with teeth. India risks becoming the jurisdiction where AI is deployed aggressively and governed weakly, a testing ground for systems too risky to run elsewhere. The digital India we are living and breathing needs evolution. The code is being written now. The conduct must follow. If we fail to govern AI with the speed and seriousness a billion citizens deserve, we will not merely fall behind. We will become a democracy where the rules are written in code by corporations, not in law by the people for the people.
