Emerging Technology and Indian Society

Mains Lens: Technology is not an external force acting on a passive society. Social institutions decide who designs it, whose data train it, which problems receive investment, who gains productivity and who bears error, surveillance, displacement and environmental cost.

Reading Guide: Build answers through technology -> social setting -> distribution of benefits and harms -> rights and power -> institutional safeguards -> human capability. Avoid both technological utopianism and technological fatalism.

Understanding Emerging Technology Socially

Emerging technologies are technologies whose capabilities, scale, applications or social rules are still developing. The category includes artificial intelligence, robotics, advanced connectivity, the Internet of Things, blockchain, drones, virtual and augmented reality, biotechnology, quantum technologies and new forms of digital infrastructure. What makes them socially significant is not novelty alone, but their capacity to reorganize knowledge, work, authority, identity, markets and everyday interaction.

Artificial intelligence is a general-purpose technology: it can be adapted across sectors rather than confined to one task. Machine learning identifies patterns from data; generative AI produces text, images, audio, video or code; robotics joins computation with physical action; and algorithmic systems rank, predict, recommend or allocate. Their output is neither pure machine objectivity nor human intention alone. It emerges from data, model design, institutional incentives, user behaviour and the context of deployment.

Concept	Meaning
Digitization	Converting analogue information into digital form, such as scanning a land record.
Digitalization	Using digital systems to reorganize an existing process, such as online applications and workflow.
Digital transformation	Redesigning institutions, roles and services around digital capability, data and new forms of participation.
Automation	Delegating a task or sequence to a machine; it may be rule-based or AI-enabled.
Augmentation	Using technology to expand human capability while preserving meaningful human judgment.
Datafication	Turning behaviour, relationships and activity into data that can be stored, analysed and acted upon.
Platformization	Organizing exchange and interaction through digital intermediaries that set rules, rank visibility and control access.

Core Distinction: Invention creates a technical possibility. Innovation turns it into a usable application. Diffusion spreads it across institutions and groups. Social transformation occurs only when norms, work, authority and opportunity change. These stages do not happen automatically or equally.

Sociological Lenses

Lens	Central idea	UPSC use
Technological determinism	Technology is treated as the independent driver of social change.	Useful as a warning about scale and speed, but it understates law, culture, markets and collective choice.
Social shaping of technology	Design and adoption reflect institutions, interests, values and user practices.	Ask who defines the problem, dataset, objective and acceptable error.
Cultural lag	Material technology may change faster than norms, law and institutions.	Explains gaps in privacy, work rules, education, consent and accountability.
Network society	Power and opportunity increasingly flow through information networks.	Connectivity creates participation, but network position and control remain unequal.
Political economy	Ownership of compute, data, platforms and intellectual property shapes distribution.	Connect innovation with concentration, rents, labour bargaining and public infrastructure.
Surveillance and discipline	Continuous visibility, scoring and prediction can influence conduct.	Use for workplaces, welfare, policing, schools, finance and platform moderation.
Capability approach	The test is whether technology expands people’s real freedoms.	Distinguish nominal access from effective, accessible and self-directed use.
Intersectionality	Technology interacts with caste, class, gender, tribe, disability, age, language and location.	The same system can include one group and deepen exclusion for another.

Social structure -> design and data -> technological system -> institutional use -> unequal outcomes -> feedback into social structure

The Emerging Technology Landscape

Technology	Social opportunity	Social risk
Artificial intelligence	Pattern recognition, prediction, generation and decision support	Bias, opacity, displacement, manipulation and concentrated compute
Robotics and automation	Precision, safety, scale and relief from hazardous tasks	Job redesign, deskilling, surveillance and unequal productivity gains
5G and Internet of Things	Connected devices, real-time monitoring and remote services	Cybersecurity, ubiquitous tracking, reliability and rural coverage
Blockchain and distributed ledgers	Tamper-evident shared records and programmable transactions	Governance errors, energy use, speculation and irreversibility
Drones	Mapping, delivery, agriculture, disaster response and inspection	Airspace safety, privacy, policing and unequal access
AR and VR	Simulation, training, cultural experience and remote collaboration	Embodied privacy, harassment, dependency and cost
Biotechnology	Health, agriculture and environmental applications	Consent, biosecurity, access, genetic discrimination and ecological uncertainty
Quantum technologies	Future advances in computing, sensing and secure communication	Cryptographic transition, strategic dependence and research concentration

These technologies increasingly converge. An agricultural drone may use satellite connectivity, AI vision, cloud computing and a data platform. A health wearable may combine sensors, predictive models and insurance decisions. Governance must therefore examine the whole socio-technical system, not regulate each component in isolation.

Technology as a Social Opportunity

Domain	Potential	Social condition
Agriculture	Localized weather and crop advice, pest detection, water management and market information	Benefit requires reliable local data, extension support, farmer choice and affordable access.
Health	Decision support, image analysis, translation, remote monitoring and assistive tools	Clinical validation, privacy, liability, continuity and human care remain essential.
Education	Translation, accessibility, teacher support, simulation and formative feedback	Technology must augment pedagogy and not widen device, language or disability gaps.
Disaster management	Risk mapping, early warning, damage assessment and resource coordination	Models express uncertainty; prediction must not replace preparedness and local knowledge.
Climate action	Grid optimization, emissions monitoring, material discovery and precision resource use	Compute, data centres, minerals and electronic waste create environmental costs.
Governance	Multilingual access, workflow support, fraud detection and improved public information	Automated exclusion, opaque scoring and weak grievance redress can injure rights.
Accessibility	Speech, vision, captioning, navigation and communication assistance	Design must work with disabled persons and preserve privacy, affordability and user control.
Small enterprise	Market discovery, design, translation, accounting and lower-cost expertise	Platform dependence, data extraction and unequal bargaining can capture gains.

Social-Good Test: A pilot is not social transformation. Ask whether the application is accurate in local conditions, affordable, accessible, trusted, supported by institutions, open to appeal and capable of reaching the last person without coercion.

Artificial Intelligence as a Socio-Technical System

AI is often described as if an autonomous model produces a neutral answer. In reality, an AI application is a chain of choices: problem definition, data collection, labelling, model design, evaluation, deployment, human interpretation and feedback. Harm can enter at any stage, and a technically accurate model can still be socially unjust if the objective itself is inappropriate.

Stage	Governance question	Typical risk
Problem framing	Who defines success, and is AI necessary?	Automating a flawed policy or measuring a proxy that ignores dignity
Data	Who is represented, absent, mislabelled or over-surveilled?	Historical discrimination, language gaps and unequal data quality
Model	What trade-offs, thresholds and error distributions are chosen?	Average accuracy hides group-specific false positives or false negatives
Interface	Can users understand, contest and use the output?	Dark patterns, automation bias and inaccessible design
Deployment	Who acts on the output and under what authority?	Function creep, coercion and use outside the validated context
Monitoring	Are outcomes, drift and unequal effects measured?	Harm persists because only system uptime is tracked
Remedy	Can a person obtain reasons, correction and human review?	Opaque decisions become practically irreversible

Purpose -> data -> model -> testing -> deployment -> human decision -> impact -> monitoring, correction and remedy

Bias, Fairness and Explainability

Bias may arise from unequal historical data, missing groups, inaccurate labels, proxy variables, model choices, interface design or the institution using the system. Removing a protected attribute does not necessarily remove discrimination because location, language, occupation or consumption may act as proxies. Fairness is also context-dependent: equal error rates, equal opportunity and equal outcomes are not always simultaneously achievable.

• Use representative and context-appropriate data, with documentation of gaps and prohibited uses.
• Evaluate performance separately across relevant social groups, languages, regions and disability conditions.
• Choose fairness criteria through public and sectoral reasoning rather than a purely technical optimization.
• Provide explanations suited to the affected person, not only technical documentation for developers.
• Retain meaningful human review, but train reviewers and prevent rubber-stamping of machine output.
• Create correction, appeal and compensation routes when automated systems cause material harm.

Work, Employment and the Platform Economy

Technology changes tasks before it changes entire occupations. Some tasks are automated; others are augmented; new tasks emerge in data preparation, maintenance, safety and human interaction. The outcome depends on adoption cost, labour institutions, market demand, worker voice, education and whether productivity gains are shared. Forecasts of a fixed number of jobs lost should therefore be treated cautiously.

Process	Opportunity	Risk
Augmentation	Decision support can raise productivity and reduce routine burden	Gains may accrue to owners while workload and monitoring intensify.
Displacement	Routine cognitive and manual tasks may shrink	Workers face income loss when transition support and job creation lag.
Task fragmentation	Platforms divide work into measurable units	Fragmentation can reduce autonomy and obscure employment responsibility.
Algorithmic management	Software allocates work, sets incentives, rates performance and disciplines workers	Opaque ratings and unilateral deactivation weaken due process and bargaining.
Skill change	Demand rises for technical, social, adaptive and domain capabilities	Short courses cannot substitute for foundational learning or real job pathways.
New hidden labour	Data annotation, content moderation and model evaluation support AI systems	Work may be low-paid, psychologically harmful or invisible in the AI value chain.
Care and relational work	Human trust, empathy, dexterity and context remain central	Low wages may persist unless social value is recognized institutionally.

Just-Transition Principle: Do not ask only whether AI creates more jobs than it removes. Ask which workers lose which tasks, who receives training and income support, who owns the productivity gain, and whether new work is dignified, secure and contestable.

A Labour-Centred Response

• Conduct sector- and task-level impact assessments with worker participation instead of relying on headline forecasts.
• Require notice, explanation and human appeal for consequential algorithmic ratings, scheduling and deactivation.
• Support lifelong learning through employers, public institutions and portable accounts while preserving broad foundational capability.
• Strengthen social security, portability and collective voice for platform and non-standard workers.
• Use public procurement and incentives to reward augmentation, safety and job quality rather than labour displacement alone.
• Share productivity gains through wages, reduced drudgery, shorter working time or wider public benefit.

The Digital Divide: Beyond Connectivity

Layer	Question
Access divide	Device, electricity, network quality, affordability and safe physical access
Skills divide	Literacy, language, digital confidence, cybersecurity awareness and ability to verify information
Usage divide	Whether technology is used for capability-enhancing purposes or only limited consumption
Outcome divide	Whether similar access produces learning, income, health or civic benefit
Design divide	Whether interfaces and datasets reflect local languages, disability needs and diverse lives
Voice divide	Whether affected communities participate in design, procurement and governance
Ownership divide	Who controls data, compute, platforms, intellectual property and resulting value

A phone shared by a household does not give each person equal access. A woman may have less time and privacy; an older person may need assisted use; a disabled person may face an inaccessible interface; a tribal-language speaker may find no usable content; and a migrant may lack stable authentication. Digital inclusion is therefore a matter of capability, agency and institutional alternatives, not device counts alone.

Technology Through an Intersectional Lens

Group	Distinctive social pathway
Women and girls	Device control, online abuse, deepfake sexualization, unpaid care, STEM pipelines, workplace culture and funding shape participation.
Caste	Historical records, residential proxies, networks and occupational patterns can encode graded inequality into apparently neutral models.
Scheduled Tribes	Connectivity, language, land data, consent and community knowledge are central; mapping can empower or dispossess.
Persons with disabilities	AI can enable communication and navigation, but inaccessible design, error and biometric exclusion can create new dependence.
Linguistic minorities	Low-resource languages receive weaker data, recognition and moderation, reducing both access and safety.
Children	Limited capacity to understand profiling, persuasive design and synthetic companionship requires heightened protection and age-appropriate design.
Older persons	Assistive tools can support independence, while scams, authentication barriers and loss of human services create vulnerability.
Rural and low-income groups	Cost, maintenance, skilled support and institutional capacity determine whether pilots become dependable services.

Artificial Intelligence and the Gender Gap

The gender gap is a pipeline, workplace, data and power problem. Equal interest or graduation does not guarantee recruitment, retention, leadership, entrepreneurial finance or control over product design. Domestic care burdens, hostile workplaces, narrow networks and biased evaluation can produce attrition at each stage.

Stage	Gender mechanism
Participation	Unequal device access, stereotypes and limited mentoring narrow entry.
Design teams	Homogeneous teams may overlook harms, contexts and user needs.
Data	Missing or distorted information about women and gender minorities affects model performance.
Deployment	Hiring, credit, safety and health systems may reproduce discrimination at scale.
Feedback	Biased outcomes become new training data and make inequality appear normal or predictive.

Unequal access -> under-representation -> data and design gaps -> biased outcomes -> reduced trust and opportunity -> continued under-representation

Beyond Representation: Diverse teams are necessary but not sufficient. Institutions also need bias testing, safe reporting, pay and promotion fairness, care-compatible work, inclusive procurement, leadership access and accountability for product harms.

Family, Culture and Identity

Institution	Transformation and tension
Relationships	Messaging and video reduce distance, but constant availability can blur boundaries and create conflict over attention and surveillance.
Parenting and childhood	Learning and creativity expand, while persuasive design, data collection, commercial influence and synthetic companionship require mediation.
Marriage and intimacy	Platforms widen choice and networks, but profiling, deception, harassment and non-consensual imagery create harm.
Culture	Digital archives and creators can revive languages and traditions, while algorithmic visibility may homogenize taste or reward stereotypes.
Identity	Online spaces enable exploration and solidarity, yet quantified popularity and permanent records can intensify comparison and stigma.
Community	Diasporic and interest-based networks build belonging, but may weaken local deliberation or produce closed publics.
Authority	Young users may gain information beyond family and community control, generating both autonomy and intergenerational tension.

Social Media and the Networked Public Sphere

Social media enables many-to-many communication through platforms that host, rank and monetize user activity. It lowers the cost of publication and collective action, but visibility is not evenly distributed. Platform rules, recommendation systems, advertising markets, influencers, organized networks and user choices jointly shape what travels.

Function	Potential social benefit
Connection	Maintains family, diaspora and community ties across distance and can reduce isolation.
Voice	Marginalized groups can document harm, find solidarity and bypass traditional gatekeepers.
Knowledge	Experts, institutions and peers can share timely educational and public-interest information.
Collective action	Rapid coordination supports relief, fundraising, accountability campaigns and civic participation.
Livelihood	Creators and small enterprises gain markets, branding, customer feedback and network access.
Culture	Local languages, humour, art and memory can circulate beyond geographic boundaries.

Risks of Platformed Communication

Risk	Social mechanism
Misinformation	False or misleading content exploits novelty, emotion and uncertainty; correction often travels more slowly.
Synthetic media	Cheap generation of realistic text, audio, image and video weakens evidentiary trust and enables impersonation.
Polarization	Ranking and selective exposure can reinforce divisions, but technology interacts with pre-existing political and social cleavages.
Hate and harassment	Scale, anonymity, group targeting and coordinated attacks impose unequal participation costs.
Non-consensual imagery	Creation or distribution of intimate and morphed content violates dignity, privacy and bodily autonomy.
Attention design	Infinite feeds, variable rewards and notifications can encourage compulsive or problematic use; effects vary by person and context.
Mental well-being	Comparison, bullying, sleep disruption and distress are risks, but association should not be written as simple universal causation.
Commercial profiling	Behavioural data supports targeted persuasion and can expose vulnerable users to manipulation or scams.
Moderation error	Under-removal leaves harm online; over-removal suppresses lawful speech, especially in low-resource languages.

Echo-Chamber Caution: Algorithms do not create polarization in a social vacuum. Identity, political competition, media systems, inequality and offline segregation also matter. A strong answer treats platform design as an amplifier and organizer, not the sole cause.

Democracy, Citizenship and Information Integrity

Digital networks expand political speech and accountability, but they also permit micro-targeting, coordinated manipulation, impersonation and rapid cross-border influence. Generative AI lowers the cost of producing persuasive content and can create a liar’s dividend: genuine evidence may be dismissed as fabricated once synthetic media becomes common.

Value	Institutional balance
Freedom of expression	Protect political criticism, satire, journalism and minority voice; vague restrictions create chilling effects.
Online safety	Act against threats, sexual abuse material, incitement, fraud and targeted harassment through lawful and proportionate process.
Information integrity	Support provenance, verification, independent research and timely correction without creating an official monopoly over truth.
Platform accountability	Require transparent rules, reasoned moderation, user notice, appeal and disclosure of systemic risk.
Electoral fairness	Disclose political advertising and synthetic campaign media and protect voters from impersonation and covert manipulation.
Pluralism	Strengthen public-interest media, local-language moderation and exposure to credible diverse sources.

Privacy, Surveillance and Data Power

Privacy is not secrecy or an obstacle to innovation. It protects dignity, autonomy, association and the ability to develop without continuous observation. Digital systems can infer sensitive traits from ordinary behaviour. Consent alone is often weak where services are essential, terms are complex, alternatives are absent or power is unequal.

Stage	Rights question
Collection	Is the data necessary for a specified, lawful purpose?
Inference	Can the system derive health, caste, religion, sexuality, politics or vulnerability from other signals?
Linkage	Does combining databases create a more intrusive profile than each dataset alone?
Retention	Is data kept beyond need, increasing breach and function-creep risk?
Sharing	Can users know which public and private actors receive data and for what purpose?
Automated action	Does a prediction affect welfare, credit, work, policing, education or health without meaningful review?
Security	Are systems resilient across vendors, supply chains, models, devices and human operators?

Governance Rule: Use necessity, proportionality, purpose limitation, data minimization, security, transparency, independent oversight and effective remedy. A technically secure surveillance system can still be unconstitutional or socially excessive.

Technology in Public Services

Digital public infrastructure and AI can lower transaction costs, improve portability and enable multilingual assistance. Yet a public system has a different duty from a commercial recommendation engine: it must preserve equality, legality, reason-giving, accessibility and remedy because denial can affect food, pension, education, liberty or livelihood.

Use	Potential	Public-law risk
Welfare	Detect duplication and guide outreach	Authentication failure, proxy discrimination and exclusion of people with weak records
Health	Triage, decision support and population planning	Sensitive data, unequal accuracy, clinical automation bias and unclear liability
Education	Translation, teacher support and formative feedback	Child profiling, vendor dependence, hallucination and reduced learner effort
Policing	Search, pattern analysis and resource support	Historical bias, false positives, mass surveillance and presumption of guilt
Justice	Research, translation and case management	Opaque prediction, confidentiality, fabricated citations and erosion of judicial reasoning
Local government	Documentation, grievance routing and multilingual access	Low capacity, poor data and centralization of local discretion

Legal authority -> defined purpose -> impact assessment -> inclusive procurement -> tested system -> human decision -> notice, appeal and audit

AI and Computational Thinking in Education

Computational thinking builds decomposition, pattern recognition, abstraction and algorithmic reasoning. Its social value is broader than coding: learners should understand how data and automated systems influence opportunity and public life. AI literacy combines use, verification, ethics, creation and civic understanding.

Dimension	Opportunity	Guardrail
Capability	Reasoning, problem-solving, creation and the ability to learn new tools	Avoid reducing future-readiness to one rapidly changing software skill.
Inclusion	Translation, captioning and assistive interfaces	Test accessibility and Indian-language performance; preserve offline pathways.
Personalization	Timely practice and feedback	Do not replace teacher judgment or intensify profiling and labelling.
Assessment	Authentic projects and process evidence	Redesign tasks rather than rely only on surveillance and detection.
Child protection	Age-appropriate exploration and safety	Minimize data, restrict manipulative design and maintain adult accountability.
Teacher role	Planning, differentiation and routine support	Professional development must prevent deskilling and automation bias.

Pedagogic Priority: Teach computational thinking before tool dependence and verification before fluency. A confident synthetic answer is not knowledge; learners need subject understanding, source evaluation, authorship norms and the courage to question machine output.

Constitutional and Judicial Architecture

Provision	Technology significance
Articles 14 and 15	Algorithmic State action must avoid arbitrariness and prohibited discrimination; formal neutrality is insufficient when outcomes are structurally unequal.
Article 19(1)(a) and 19(2)	Online speech and access to information receive constitutional protection, subject to legally authorized and reasonable restrictions on listed grounds.
Article 19(1)(g) and 19(6)	Digital occupations and enterprises are protected, subject to reasonable restrictions in the public interest.
Article 21	Life and personal liberty support dignity, autonomy, privacy and fair procedure in data-intensive systems.
Articles 38 and 39	The State should reduce inequality and prevent concentration of wealth and means of production; relevant to platform and data power.
Article 41	Education, work and public assistance connect technology policy with capability and transition support.
Article 51A(h)	Scientific temper, humanism and inquiry require both innovation and critical evaluation of technological claims.
Article 51A(e) and (g)	Dignity of women and environmental responsibility are relevant to online abuse, biased systems and technology’s material footprint.

Judicial Anchors

Case	Usable principle
Shreya Singhal (2015)	Struck down Section 66A of the information-technology law for vagueness and overbreadth; distinguished discussion and advocacy from incitement.
K.S. Puttaswamy (2017)	Recognized privacy as a fundamental right grounded in dignity and liberty; State intrusion requires legality, legitimate purpose, proportionality and safeguards.
Anuradha Bhasin (2020)	Protected speech and trade through the internet as constitutional activity; indefinite suspension is impermissible and restriction orders require publication and review.

These principles do not imply an absolute right to any platform or technology. They require lawful authority, precise purposes, proportional restrictions, procedural safeguards and remedy when the State or regulated intermediaries affect protected interests.

A Risk-Based Governance Framework

Not every AI use requires the same control. Spell-checking, medical triage, welfare eligibility and autonomous weapons present different stakes. Governance should be proportional to severity, scale, reversibility, vulnerability, autonomy, uncertainty and the availability of alternatives.

Risk level	Typical character	Proportionate response
Minimal	Low-stakes assistance with little effect on rights	Basic transparency, security and user control
Limited	Recommendation or content generation with manageable harm	Disclosure, provenance, testing and complaint channels
High	Health, employment, education, credit, welfare, policing or biometric decisions	Impact assessment, strong validation, group testing, human review, logs, audit and appeal
Intolerable	Uses incompatible with dignity, legality or meaningful freedom	Prohibition or strict non-deployment

Principles for Responsible Technology

Principle	Operational meaning
Legality	A clear legal basis and defined institutional responsibility
Necessity	Demonstrate that technology addresses a real problem and that a less intrusive option is inadequate
Proportionality	Match data, automation and restriction to the importance and risk of the objective
Fairness	Test unequal errors, exclusion and accessibility across relevant groups
Transparency	Disclose automated use, purpose, limits and decision pathways in understandable form
Accountability	Identify a responsible human and institution; a vendor contract cannot absorb public duty
Contestability	Provide notice, reasons, correction, human review and timely remedy
Security and resilience	Protect the full lifecycle, supply chain, model, data and operational context
Sustainability	Measure energy, water, minerals, hardware life and electronic waste
Participation	Include affected communities, workers and independent experts before and after deployment

Current Status (as of August 2026)

Area	Position
IndiaAI Mission	The Mission was approved in March 2024 with an outlay of Rs 10,371 crore over five years. Its seven pillars cover compute, foundation models, datasets, applications, future skills, startup finance and safe and trusted AI.
Mission progress	By late July 2026, official reporting identified 20 indigenous model proposals for support, 15 empanelled compute providers, 237 supported compute projects and 93 lakh sanctioned GPU-hours. These are programme counts, not measures of social impact.
Applications and talent	The same July 2026 update reported 62 AI prototypes, 20 deployed solutions, 686 fellowships across 178 institutions and more than 26 lakh completions under YUVA AI for All. Completion does not by itself establish skill mastery or employment.
Distributed capacity	Twenty-seven India Data and AI Labs had been established, with work continuing on 188 more; 58 AI Centres of Excellence were being established with States, Union Territories and industry partners.
Safe and trusted AI	Thirteen responsible-AI projects had been approved in areas including bias mitigation, machine unlearning, privacy, explainability, deepfake detection and risk assessment. The AI Safety Institute had been established through a hub-and-spoke model.

Current Status (as of August 2026): These figures are frozen at August 2026 and will change. Use them to show movement from an AI strategy based mainly on aspiration towards one involving shared compute, domestic models, skilling and safety institutions – while still asking who can access these resources and how outcomes are evaluated.

Area	Current position
AI governance	Final India AI Governance Guidelines were released in November 2025. They adopt a principle-based, risk-sensitive and techno-legal approach, relying mainly on existing laws and sector regulators rather than a new horizontal AI statute at this stage.
Institutional coordination	The AI Governance and Economic Group was constituted in April 2026 as the apex inter-ministerial coordination body, supported by a Technology and Policy Expert Committee.
Personal data	The Digital Personal Data Protection Rules, 2025 were notified in November 2025 with phased implementation extending up to eighteen months. The framework strengthens notice, consent, security, breach communication and safeguards for children, subject to its terms and exemptions.
Synthetic media	The intermediary rules were amended in February 2026 to address synthetically generated information. Significant social-media intermediaries must use user declarations, reasonable verification measures and prominent labels under the applicable provisions.
School curriculum	For the 2026-27 session, the launched curriculum is specifically a CBSE computational-thinking and AI curriculum for Classes III to VIII. It should not be described as proof that every school board and every school in India has already implemented an identical curriculum.
Cybercrime response	The national cybercrime coordination and reporting architecture, cyber incident response and botnet-cleaning functions remain distinct. The botnet-cleaning centre is not a dedicated social-media moderation or malicious-actor attribution agency.

Current Status (as of August 2026): The legal position is an evolving layered framework: constitutional rights, criminal and civil law, personal-data rules, intermediary duties, consumer and intellectual-property law, sector regulation, technical standards and non-binding AI governance guidance. Do not write that India has no data-protection framework or, at the other extreme, that it has one comprehensive AI Act.

Way Forward: A Human-Centred Technology Compact

Build Universal Digital Capability

• Treat affordable connectivity, electricity, accessible devices, public access points and local support as basic capability infrastructure.
• Invest in Indian-language datasets, interfaces and moderation with community participation and quality evaluation.
• Make accessibility a procurement and testing requirement from the beginning, not a later accommodation.
• Teach digital, media, data and AI literacy across the life course, including verification, privacy, cybersecurity and rights.

Govern High-Risk Systems Across Their Lifecycle

• Require algorithmic or technology impact assessments before high-risk public and workplace deployment.
• Document purpose, data, model limitations, affected groups, fallback arrangements and prohibited secondary uses.
• Use independent testing, red-teaming and group-disaggregated performance measures suited to Indian contexts.
• Monitor drift and real-world outcomes after deployment; suspend systems when harm cannot be corrected.

Protect Rights and Preserve Human Agency

• Guarantee notice, understandable reasons, correction, human review and time-bound appeal for consequential automated decisions.
• Retain offline and assisted alternatives for essential services so that technology does not become coerced access.
• Limit biometric and behavioural surveillance through legality, necessity, proportionality and independent oversight.
• Create special protections for children, workers, disabled persons and communities subject to concentrated monitoring.

Make Platforms Accountable Without Weakening Liberty

• Require clear content rules, user notice, reasoned action, accessible appeals and transparency about systemic moderation risk.
• Strengthen provenance and labelling of synthetic media while supporting independent verification and public-interest research.
• Invest in local-language moderation and survivor-centred response to harassment and non-consensual imagery.
• Avoid universal mandatory identity verification; accountability measures must also protect privacy, anonymity and vulnerable speech.

Distribute Innovation and Productivity

• Treat shared compute, public-interest datasets and open standards as infrastructure accessible to academia, startups and social innovators.
• Use competition policy, interoperability and data portability to reduce lock-in and excessive platform concentration.
• Support workers through transition finance, portable social security, collective voice and credible pathways from training to jobs.
• Evaluate public funding by social outcomes, regional diffusion, accessibility and environmental cost, not only patents or model size.

Reduce the Material Footprint

• Disclose energy, water and hardware impacts for large computing projects and locate infrastructure within resource constraints.
• Improve model and data-centre efficiency, renewable-energy matching, heat reuse where feasible and responsible water management.
• Extend device life through repairability, procurement standards, recycling and formal protection for e-waste workers.
• Assess rebound effects: efficiency can lower unit cost but increase total use.

Access and capability + responsible design + risk-based oversight + worker transition + platform accountability + remedy -> technology that enlarges freedom

Guiding Principle: India should move from AI for scale to AI for accountable capability. The decisive question is not whether a system is innovative, but whether it expands human agency while distributing benefit, error, voice and remedy fairly.

Mains Answer Toolkit

Ten-Mark Structure

• Define emerging technology as a socio-technical system rather than an autonomous force.
• Identify one opportunity and two distributional risks across access, work, rights or identity.
• Use one constitutional principle or sociological lens and one institutional safeguard.
• Conclude with augmentation, capability and contestability.

Fifteen-Mark Structure

• Introduction: technology reorganizes social relations; it does not act outside society.
• Concept: social shaping, cultural lag, network society, datafication and platform power.
• Opportunity: health, agriculture, education, climate, governance and accessibility.
• Inequality: class, caste, tribe, gender, disability, age, language, region and ownership.
• Institutions: work, family, public sphere, privacy, welfare and public services.
• Governance: constitutional rights, risk tiers, lifecycle controls, human review and remedy.
• Evidence: one dated item from the August 2026 status block.
• Conclusion: democratize capability, distribute gains and preserve human agency.

Reusable Introductions and Conclusions

Introduction: Emerging technology does not merely add new tools to society; it redistributes visibility, knowledge, bargaining power and the capacity to decide. Its social value therefore depends on institutions that convert innovation into shared capability rather than concentrated control.

Conclusion: A democratic technology order must combine innovation with rights, scale with accessibility, automation with worker security, data with dignity and platform power with contestability. Human beings must remain the authors and beneficiaries of technological change.