Content :

1. AI & Copyright Law
2. HCES 2023–24
3. Saturday classes, home lessons: Mizoram road to ‘first fully literate state’
4. AI in India: strategy must precede mission
5. Are gig workers a part of India’s labour data?

---

AI & Copyright Law

---

Core Issue

1. At the heart of the debate: Are generative AI models built on copyrighted works?
2. Two major US copyright lawsuits (Writers vs. Anthropic & Authors vs. Meta) question whether AI training violates copyright law.

Source : The Indian Express

Relevance : GS 2(Governance ) ,Gs 3(IPR , Technology)

Court Verdicts So Far

1. Writers vs. Anthropic (Aug 2023, US)

1. Authors including Michael Chabon, George R.R. Martin filed case.
2. Accused Anthropic of copying copyrighted texts for training Claude AI.
3. Court’s response: Did not rule on copyright infringement directly; stated the AI model does not “transform” texts enough to qualify as fair use.

2. Authors vs. Meta

1. Authors sued Meta for training LLaMA models using their copyrighted books.
2. Judge dismissed part of the complaint on procedural grounds.
3. However: Meta could still be held liable if models “memorise and regurgitate” copyrighted content.

Key Legal Concepts

1. Fair Use: Permits limited use of copyrighted work without permission if it transforms the content (e.g. parody, research).
2. Transformative Use: AI must add new expression or meaning to qualify.

The Case Against OpenAI in India

1. In 2024, ANI and the Indian Music Industry (IMI) accused OpenAI of:
   1. Training models on copyrighted Indian content.
   1. Violating Section 65A of Indian Copyright Act (on circumvention of tech protections).
2. No judgment yet. Jurisdiction under question.

Challenges in India

1. OpenAI operates in India, but:
   1. No clarity on how training data is sourced.
   1. AI firms claim only “public” data is used.
   1. Lack of explicit Indian law on AI & copyright.

Why It Matters

1. India’s creative industry (books, music, cinema) is at risk of unauthorised AI replication.
2. Worries about AI models “memorising and regurgitating” original work.
3. Raises ethical & legal questions around ownership and consent.

Global Implications

1. Courts have not yet conclusively ruled if using copyrighted work for AI training is legal.
2. Verdicts will set precedents for AI governance globally, affecting OpenAI, Meta, Google, Anthropic, etc.

Significance of Rulings

1. So far, courts have favoured tech companies but have not shut the door on future liability.
2. If plaintiffs prove “verbatim memorisation” by models, it could trigger compensation or licensing models.

Key UPSC-Relevant Facts from the Article

Fact	UPSC Relevance
Over 7 million pirated books allegedly used to train Anthropic’s Claude AI	Raises ethical and legal concerns over copyright violations
US Courts ruled that using books for AI training can qualify as “transformative fair use”	Insight into evolving jurisprudence in digital IPR – may influence Indian legal reforms
Meta’s case dismissed for lack of proof of harm, not because AI use is legal	Illustrates complexities in proving “economic harm” in copyright law
In India, OpenAI has no direct data center or formal license for copyright-covered models	Reflects gaps in India’s digital regulatory framework for GenAI models
India-Germany Joint Declaration (2022) on Triangular Cooperation (TrC)	Related to India’s role in shaping global tech governance (GS II/IR angle)
India has no settled law on AI and copyright yet	Opportunity for reform under Digital India Act or IPR amendments

Key Dimensions:

Legal Gaps in India

• No AI-specific copyright law.
  • Copyright Act, 1957 doesn’t define AI authorship or fair use for training data.

Ethical Concerns

• Use of creative content without consent or credit.
  • Undermines originality and creator rights.

Economic Impact

• Threat to livelihoods of artists, authors, musicians.
  • Monetization of pirated or public content by AI companies.

Technology vs Regulation

• Balance between fostering innovation and protecting IP.
  • Ambiguity over “transformative use” of copyrighted material.

Global Comparisons

• US: “Fair use” doctrine allows AI training.
  • EU: Tight opt-outs under TDM rules.
  • Japan: Broad AI training exemptions.

Regulatory Vacuum

• No guidelines under IT Act or DPDP Act for AI training data.
  • Digital India Act still pending.

Privacy and Consent

• Training data may include personal content without consent.
  • Conflicts with data protection principles.

Creator Rights & Royalties

• Lack of collective bargaining tools (e.g., CMOs for AI usage).
  • No attribution mechanism for original creators.

AI Liability & Accountability

• Who is responsible for AI-generated infringements — developer or deployer?
  • No legal clarity yet.

Public Good vs Private Profit

• Use of public domain data for private AI profit.
  • Debate over open-source mandates for public-trained models.

---

HCES 2023–24

---

What is HCES?

• HCES = Household Consumption and Expenditure Survey
• Conducted periodically by the National Sample Survey Office (NSSO) under the Ministry of Statistics and Programme Implementation (MoSPI).
• It captures detailed data on household consumption patterns, income, and living standards across rural and urban India.

Source : The Indian Express

Relevance : GS 2(Governance , Social Issues)

What does the HCES 2023–24 show?

• Reports per capita daily calorie, fat, and protein intake.
• Provides insights into nutritional status, consumption inequality, and shifts in food habits.
• Compares data across different income deciles, helping track changes among the top 5% vs bottom 5%.
• First such release after over a decade (since 2011–12 round), delayed due to COVID-19 disruptions.

Key Findings from HCES 2023–24 (Nutritional Intake Data)

Indicator	2011–12	2022–23	Change
Daily Calorie Intake (Rural)	2,233 kcal	2,212 kcal	↓ 0.94%
Daily Calorie Intake (Urban)	2,240 kcal	2,230 kcal	↓ 0.4%
Protein Intake (Rural)	61.9 g	61.8 g	~Stable
Protein Intake (Urban)	63.2 g	63.4 g	↑ Slight
Fat Intake (Rural)	59.7 g	60.4 g	↑ 1.17%
Fat Intake (Urban)	70.5 g	69.8 g	↓ 1.0%
Bottom 5% Calorie Intake (Rural)	1,607 kcal	1,688 kcal	↑ 5%
Top 5% Calorie Intake (Rural)	3,116 kcal	2,941 kcal	↓ 5.6%
Bottom 5% Calorie Intake (Urban)	1,623 kcal	1,696 kcal	↑ 4.5%
Top 5% Calorie Intake (Urban)	3,478 kcal	3,092 kcal	↓ 11%

Insights & Relevance for UPSC

• Calorie Inequality Down: Significant narrowing between top 5% and bottom 5% across rural and urban areas.
• Slight Calorie Dip: Overall calorie intake down marginally but protein remains stable—indicating changing food preferences.
• Policy Relevance:
  • Targets for schemes like POSHAN Abhiyaan, NFSA, Mid-Day Meal, and PM Garib Kalyan Yojana.
  • Evidence for SDG-2 (“Zero Hunger”) progress.
• Health Implication: Drop in fat intake among richer groups suggests growing health awareness.

Additional Dimensions to Cover

1. Link to SDGs

• SDG 2: End hunger, achieve food security and improved nutrition.
• The narrowing gap supports Target 2.1 (access to food) and 2.2 (end all forms of malnutrition).

2. Inequality & Welfare Economics

• Reflects reduced nutritional inequality, possibly due to schemes like NFSA, POSHAN Abhiyaan, and PM-GKAY.
• Suggests improved welfare targeting of subsidies and rations.

3. Urban-Rural Nutrition Divide

• Urban nutrition is more stable; rural decline in calorie intake needs further analysis — is it due to underconsumption or dietary transition?

4. Food Security vs. Nutrition Security

• India is shifting from calorie sufficiency to nutritional adequacy.
• Calorie intake decline might mask hidden hunger (micronutrient deficiency).

5. Behavioural & Cultural Shifts

• Decline in fat intake and calorie-rich foods by the top 5% indicates rising health awareness, lifestyle diseases focus, and shift to balanced diets.

6. Role of Inflation and Food Prices

• Rising food prices (especially pulses, oils, proteins) may have reduced consumption among poor, even if calorie inequality narrowed.

7. Data Limitations

• HCES data may underreport consumption of processed foods or dining out.
• Calorie data doesn’t capture micronutrient adequacy or meal diversity.

8. Gender and Age-Based Access

• No disaggregated data provided — intra-household disparities (e.g., women, elderly, children) still a concern.

---

Saturday classes, home lessons: Mizoram road to ‘first fully literate state’

---

Context & Milestone

• Mizoram has become India’s first fully literate state under the ULLAS (Nav Bharat Saksharta Karyakram).
• Achieved 98.2% literacy for population aged 7+ (PLFS 2023–24); surpasses Kerala (96.2%).

Relevance : GS 2(Education, Governance)

What is ULLAS?

• Union Government’s flagship adult literacy mission launched in 2022.
• Targets non-literate people aged 15+ with a 5-year timeline.
• Aims at foundational literacy, numeracy, and digital & financial literacy.

Key Drivers of Mizoram’s Success

• Community-driven efforts and strong volunteer participation.
• Adult learners (like 94-year-old Latinkimi) attend classes before/after farming.
• Localised teaching materials developed historically by missionaries (Mizo language primers, textbooks).

Historical Literacy Trends

• 1991: Kerala declared ‘totally literate’ with 90% adult literacy (NLM norms).
• 2011 Census: Kerala – 93.91%, Mizoram – 91.58%.
• PLFS 2023–24: Mizoram – 98.2% (7+ years), Kerala – 96.2%.

Why It Matters

• Sets a national benchmark in grassroots adult literacy.
• Aligns with SDG 4 (Quality Education) and India’s broader Digital and Inclusive Education goals.
• Emphasises lifelong learning, not just school-based literacy.

Historical & Comparative Data

Indicator	Value
India’s adult illiterate (15+)	~15 crore (MoE, 2022)
National Literacy Rate	~77.7% (PLFS 2022–23)
1991 Kerala Adult Literacy	90% (declared ‘total literacy’ by NLM norms)
2011 Census	Mizoram: 91.58%

India’s Literacy Rates – Census 2011

Category	Literacy Rate (%)
Overall (National Average)	74.04%
Male	82.14%
Female	65.46%
Urban Areas	84.11%
Rural Areas	67.77%

---

AI in India: strategy must precede mission

---

Context & Why in News

• India aspires to be a global leader in AI governance, positioning itself as a voice for the Global South.
• However, it lacks a comprehensive, democratically anchored National AI Strategy, risking technocratic and opaque governance.

Relevance : GS 3(Technology ) ,GS 2(Governance)

Key Issues Identified

• No National AI Strategy: Current approach via IndiaAI Mission is implementation-focused, not strategic.
• Lack of National Priorities: Unclear values, governance structures, and sectoral priorities.
• Opaque Governance: Centralisation without parliamentary oversight or civil society input.
• Technological Dependency: Absence of indigenous strategy may lead to strategic reliance on foreign AI systems.

Major Risks & Gaps

• Strategic Autonomy: AI used in defence/intelligence makes sovereignty critical.
• Employment Disruption:
  • 65,000 IT jobs lost in 2024 (TCS, Infosys, Wipro).
  • IMF: 26% of India’s workforce exposed to GenAI, 12% at risk of displacement.
  • No plan for reskilling, labour transition, or social protection.
• Environmental Cost:
  • AI’s energy needs rising.
  • Data centres stress power & water in cities like Bengaluru, Hyderabad.
• Social Impact Unaddressed:
  • AI in health, welfare, policing may amplify bias, hurt accountability.
  • No inclusive policy on AI ethics, equity, or digital rights.

What Should Be Done? (Recommendations)

1. Publish a National AI Strategy endorsed by the Cabinet and table it in Parliament.
2. Form a Standing Parliamentary Committee on AI & Emerging Tech.
3. Commission AI Employment Impact Study: Granular insights on job loss by sector, region, demographic.
4. Build institutional architecture via public consultations and democratic dialogue.
5. Align AI with national security, economic resilience, equity, and sustainability.

Global Implications

• Without internal coherence, India’s global leadership in AI forums like Global Partnership on AI will lack credibility.
• India must model democratic AI governance for the developing world.

Data & Statistics

Data Point	Figure	Context / Source
IndiaAI Mission budget	₹10,371 crore	Announced in Union Budget 2024–25
Job losses (2024)	~65,000 jobs	Lost across TCS, Infosys, Wipro,etc
IMF Estimate – Workforce at risk	26% of Indian workforce exposed to GenAI	12% of jobs at risk of displacement
Energy demand forecast (Global)	Will double by 2030	Source: International Energy Agency
Water stress	11 out of India’s top 20 cities	AI hubs like Bengaluru & Hyderabad face severe groundwater decline
Adult literacy needed for ‘total literacy’ under NLM (for earlier reference)	90% aged 15–35	For comparison: Kerala achieved this in 1991

Key Institutional Mentions

• IndiaAI Mission: Implemented under Section 8 company of MeitY.
• Global Partnership on AI (GPAI): India has leadership role.
• Future of India Foundation: Publisher of the source report, “Governing AI in India: Why Strategy Must Precede Mission”.

---

Are gig workers a part of India’s labour data?

---

Context & Why in News

• Union Budget 2025 formally recognised gig & platform workers and extended some welfare benefits.
• But the PLFS 2025 (Periodic Labour Force Survey) fails to statistically identify gig/platform workers, undermining data-driven policymaking.

Relevance : GS 2(Governance , Labour Welfare)

Gaps in Labour Classification

• Gig workers defined in the Code on Social Security, 2020, but ambiguously.
• PLFS still classifies them under ‘self-employed’, ‘casual labour’, or ‘own-account worker’ — masking their unique conditions.
• Algorithmic work, multi-platform juggling, no job security, and app-based tasking ≠ traditional employment.

Consequences of Statistical Invisibility

• Gig workers lack social security representation despite legal recognition.
• Policy efforts (e-Shram portal, AB-PMJAY, digital ID cards) lack strong data support.
• Welfare boards rely on PLFS data — flawed classification → exclusionary access to schemes.

Why Gig Work Is Different

• No stable contract or control over work → cannot be equated with self-employment.
• Shaped by algorithms, task-based pay, platform switching.
• Employment volatility, digital reach dependence, and zero benefits not reflected in PLFS.

Policy & Data Disconnect

• MoSPI response in Rajya Sabha: Gig workers are included under ‘economic activity’ — but not separately identified.
• PLFS 2025 changes: More rural representation, monthly estimates — but no changes to capture gig-specific variables.

Way Forward

• Update PLFS classification codes to include gig/platform categories.
• Introduce special survey modules to assess:
  • Number of platforms worked on
  • Nature of algorithmic control
  • Work hours, contracts, pay volatility
• Institutionalise National Social Security Fund (Clause 141) and Welfare Boards (Section 6 of Code) with gig-specific inputs.

Relevant Data Points

• 23.5 million gig workers projected by 2029–30 (NITI Aayog 2022).
• Despite inclusion in Code on Social Security, gig workers remain invisible in core labour statistics.

Additional Dimensions

1. Legal and Policy Recognition

• Code on Social Security, 2020: Defines gig and platform workers but lacks operational clarity.
• No corresponding rules/guidelines yet framed to implement social security benefits.

2. Statistical Gaps

• PLFS 2025: Fails to create a distinct category for gig workers.
• Gig workers merged into ‘self-employed’, ‘casual labour’, causing data invisibility.

3. Economic and Social Security Issues

• No written contracts, volatile incomes, no maternity benefits, or insurance by default.
• e-Shram portal (~30 Cr+ registrations) lacks granularity for gig-specific insights.

4. Algorithmic Control and Digital Governance

• Gig work is governed by platform algorithms, which determine work allocation, pay, and reviews — often without transparency.
• Raises issues of digital rights, data privacy, and worker surveillance.

5. Labour Rights and Unionisation

• Gig workers are not covered under Trade Union Act, limiting collective bargaining.
• Attempts at forming informal unions (e.g., Swiggy delivery agents’ strikes).

6. Gender and Gig Economy

• Low female participation in platform work due to safety, tech access, and lack of flexibility.
• Women gig workers face wage gaps and higher care burden.

7. International Comparisons

• EU passed legislation mandating algorithmic transparency for platform workers.
• UK Supreme Court (2021): Ruled Uber drivers are ‘workers’, not self-employed.

Additional Data and Facts:

Indicator	Data Point	Source
Projected gig workforce	23.5 million by 2029–30	NITI Aayog Report (2022)
Current estimate	7.7 million (as of 2020–21)	NITI Aayog
Women in gig economy	~10% of gig workforce	NITI Aayog
e-Shram registrations	30 crore+	Ministry of Labour & Employment
PLFS 2022-23 gig data	Not separately identified	MoSPI response in Parliament
Informal sector coverage	92.4% of total workforce (2022)	PLFS 2022-23
Workers without written contract	Over 70% of informal workers	PLFS & ISLE

Other angles to explore:

• Urbanisation and gig clustering (metro-centric platforms).
• Impact of AI & automation on gig jobs (task elimination, micro-tasking).
• Sustainability concerns — frequent travel by delivery/rideshare gig workers.
• Social audits & grievance redressal in gig platforms — almost non-existent.
• Need for a Unified Labour Market Framework inclusive of informal and gig workers.