AI & Digital Ethics Grade 9 AI in India Applications and Policy

Study Guide: AI in India – Applications and Policy
Grade 9 | Social Studies & Digital Ethics

1. The Driving Question

"India is using AI to solve problems like crop failures, traffic jams, and fake news—but who decides which problems get solved, and what happens when the AI gets it wrong? If a farmer’s loan is denied by an algorithm or a hospital’s AI misdiagnoses a patient, can they challenge the decision, or is the code the final judge?"

2. The Core Idea – Built, Not Listed

Imagine you’re in Pune, India, during monsoon season. A farmer named Rajesh checks his phone: an AI app called Kisan AI predicts his cotton crop will fail in two weeks due to pests. The app suggests spraying a specific pesticide—but Rajesh can’t afford it. Meanwhile, in Mumbai, a traffic camera uses AI to spot a stolen car, but the system flags the wrong license plate, causing a police chase for an innocent driver. In Delhi, a hospital’s AI tool scans X-rays and tells a doctor a patient has tuberculosis—but the doctor disagrees and orders a second test, which comes back negative.

These aren’t sci-fi scenarios; they’re real AI systems being used right now in India to make decisions that affect people’s lives. But here’s the puzzle: Who is responsible when the AI is wrong? Is it the government that approved the system? The company that built it? The farmer who trusted the app? Or the code itself? India’s approach to AI is a balancing act—using technology to leapfrog old problems (like poverty or poor infrastructure) while trying to prevent new ones (like bias, job loss, or unaccountable decisions). Unlike the U.S. or China, India doesn’t have a single "AI law." Instead, it relies on a mix of policy guidelines, court rulings, and public pressure to shape how AI is used.

Key Vocabulary:
- Algorithm Bias
Definition: When an AI system produces unfair outcomes because the data it was trained on reflects existing inequalities.
Example: An AI hiring tool in India might favor candidates from elite universities because its training data overrepresented them—even if those schools aren’t the best predictor of job success.
College Note: In advanced ethics courses, bias is studied as a structural issue (e.g., how colonial-era data shapes modern AI) rather than just a technical glitch.

• Digital Public Infrastructure (DPI)
  Definition: Government-built digital systems (like India’s Aadhaar ID or UPI payments) that private companies can use to build AI tools.
  Example: A startup in Bengaluru uses Aadhaar data to verify identities for a loan app, but if the data is outdated, the AI might deny loans to people who need them.
  College Note: DPI is debated in global policy—some see it as a way to democratize tech, others as a tool for state surveillance.

• Explainable AI (XAI)
  Definition: AI systems designed to show how they reached a decision, so humans can challenge or understand it.
  Example: If an AI denies someone a government subsidy, an XAI system might say, "Denied because your income data from 2020 was flagged as inconsistent with your 2022 tax filings." College Note: XAI is a growing field in computer science, but critics argue it’s often used to justify decisions rather than make them fairer.

• Jugaad Innovation
  Definition: A Hindi term for creative, low-cost solutions to problems—often used to describe how India adapts AI with limited resources.
  Example: Farmers in Karnataka use WhatsApp groups to crowdsource pest alerts when they can’t afford AI tools, showing how "low-tech" solutions compete with high-tech ones.
  College Note: Jugaad is studied in business schools as a model for frugal innovation, but it also raises questions about sustainability and scalability.

3. Assessment Translation

Format: This topic appears in Grade 9 Social Studies as: - Short-answer questions (e.g., "Explain one way AI in India could reduce inequality and one way it could increase it.") - Document-based questions (DBQs) (e.g., analyzing a news article about AI in healthcare alongside India’s National Strategy for AI policy) - Debate prompts (e.g., "Should India prioritize AI for economic growth or for social welfare?") - SAT/ACT Connection: While not directly tested, the critical reading and evidence-based writing skills apply to passages about technology and ethics (e.g., analyzing an author’s argument about AI bias).

What a Proficient Response Looks Like:
- Developing: "AI helps farmers by predicting weather. It can also be bad because it might make mistakes." - Proficient: "AI tools like Kisan AI help farmers by predicting crop failures, which can reduce food waste and increase incomes. However, if the AI is trained on data from wealthy farms, it might give bad advice to small farmers who can’t afford the suggested solutions. India’s National Strategy for AI tries to address this by promoting ‘AI for All,’ but critics argue the policy doesn’t do enough to regulate private companies that build these tools. For example, if a farmer’s loan is denied by an AI, there’s no clear way to appeal the decision, which could deepen inequality."

Model Student Response (Proficient):
Prompt: "How does India’s approach to AI policy differ from that of the U.S. or China? Use one example from each country to support your answer."

"India’s AI policy is more reactive than the U.S. or China because it focuses on solving immediate problems rather than long-term dominance. For example, India’s National Strategy for AI (2018) emphasizes ‘AI for Social Good,’ like using AI to detect tuberculosis in rural clinics. In contrast, the U.S. relies on private companies (like Google or Microsoft) to set AI standards, which can lead to gaps in regulation—like when an AI hiring tool was found to discriminate against women. China, meanwhile, uses AI for state surveillance, like its ‘social credit’ system, which tracks citizens’ behavior. India’s approach is closer to the U.S. in its reliance on private innovation, but it lacks China’s centralized control. This creates a tension: India wants to use AI to improve lives, but without strong laws, companies might prioritize profit over fairness, like when an AI loan app denied credit to low-income borrowers without explanation."

What Teachers Look For:
- Evidence: Specific examples (e.g., Kisan AI, Aadhaar, National Strategy for AI).
- Analysis: Explaining why a policy matters (e.g., "India’s lack of a data protection law means companies can use personal data without consent").
- Nuance: Acknowledging trade-offs (e.g., "AI can reduce traffic deaths but might also lead to job losses for drivers").

4. Mistake Taxonomy

Mistake 1: Overgeneralizing AI’s Impact
Prompt: "Describe one benefit and one risk of using AI in Indian agriculture." Common Wrong Response: "AI helps farmers grow more food, but it can also take away their jobs." Why It Loses Credit: - Vague: "More food" and "take away jobs" are too broad—what kind of food? Which jobs? - No Evidence: Doesn’t name a specific AI tool or policy.
Correct Approach: 1. Name the tool: "AI tools like Kisan AI help farmers by predicting pest outbreaks, which can save crops like cotton or wheat." 2. Explain the benefit: "This reduces the need for expensive pesticides, lowering costs for small farmers." 3. Name the risk: "However, if the AI is trained on data from large farms, it might recommend solutions (like certain seeds or fertilizers) that small farmers can’t afford, widening the gap between rich and poor farmers." 4. Connect to policy: "India’s National Strategy for AI encourages ‘AI for All,’ but without regulations, private companies might not address this bias."

Mistake 2: Ignoring the Human Role in AI Decisions
Prompt: "If an AI system denies a farmer a loan, who is responsible—the government, the company, or the farmer? Explain your answer." Common Wrong Response: "The company is responsible because they made the AI." Why It Loses Credit: - Oversimplified: Doesn’t consider how policies or data shape the AI’s decisions.
- No Nuance: Doesn’t acknowledge shared responsibility.
Correct Approach: 1. Company: "The company is responsible for building the AI, but they might argue the system is ‘neutral’ because it’s based on data." 2. Government: "The government is responsible for regulating how companies use data—if there’s no law requiring explainable AI, the farmer can’t challenge the decision." 3. Farmer: "The farmer might share responsibility if they provided incorrect data, but this raises ethical questions: Should a farmer be punished for a mistake in a system they don’t understand?" 4. Policy Gap: "India’s Digital Personal Data Protection Act (2023) gives citizens the right to know how their data is used, but it doesn’t require companies to explain AI decisions, leaving farmers with little recourse."

Mistake 3: Confusing AI with Automation
Prompt: "How might AI change the job market in India? Give one example of a job that could be created and one that could be lost." Common Wrong Response: "AI will take all the jobs, like drivers and factory workers. It will create jobs for programmers." Why It Loses Credit: - Catastrophizing: "Take all the jobs" is an exaggeration.
- No Specificity: Doesn’t name real industries or AI tools.
Correct Approach: 1. Job Lost: "AI-powered self-checkout kiosks in stores like Reliance Retail could reduce the need for cashiers, especially in urban areas where labor costs are higher." 2. Job Created: "AI could create jobs for ‘AI trainers’—people who label data to improve systems, like teaching an AI to recognize local dialects in customer service chatbots." 3. Policy Context: "India’s Skill India Mission is trying to retrain workers for AI-related jobs, but critics argue the program doesn’t move fast enough to match the pace of automation." 4. Jugaad Example: "In rural areas, AI might not replace jobs but augment them—like using AI to help ASHA workers (community health workers) diagnose diseases, allowing them to serve more patients."

5. Connection Layer

1. Within Social Studies: AI in India → India’s Federalism

2. Why it matters: India’s states have different AI priorities—Kerala focuses on AI in education, while Telangana pushes for smart cities. Understanding AI policy helps explain why some states resist or adopt central government tech initiatives (like Aadhaar), revealing tensions in India’s federal structure.

3. Across Subjects: AI Bias → Statistics (Math)

4. Why it matters: AI bias often stems from sampling errors in training data. For example, if an AI loan app is trained mostly on data from urban borrowers, it might unfairly reject rural applicants—a problem that becomes clearer when you study confounding variables in statistics.

5. Outside School: AI in India → Bollywood Deepfakes

6. Why it matters: In 2023, a deepfake video of Indian actor Rashmika Mandanna went viral, sparking debates about AI’s role in misinformation. Now, when you see a viral video, you’ll ask: Is this real, or is it an AI-generated ‘hallucination’? This connects to India’s IT Rules (2021), which require social media platforms to label deepfakes—but enforcement is weak.

6. The Stretch Question

"India’s Aadhaar system gives every citizen a unique digital ID, which is used for everything from bank accounts to welfare benefits. Some argue Aadhaar is a tool for financial inclusion, while others say it’s a form of digital surveillance. If an AI system uses Aadhaar data to decide who gets a loan or a government subsidy, is that empowerment or control? Where would you draw the line—and who should get to draw it?"

Pointer Toward an Answer:
- Empowerment Argument: Aadhaar + AI can reduce corruption by ensuring subsidies go to the right people (e.g., stopping "ghost beneficiaries" who don’t exist). For example, Direct Benefit Transfers (DBT) use Aadhaar to send money directly to farmers, cutting out middlemen.
- Control Argument: If the government or companies can track how you use your Aadhaar (e.g., where you spend your money, which hospitals you visit), it could be used to profile citizens. For example, if an AI loan app denies you credit because your Aadhaar data shows you shop at "low-income" stores, is that fair? - The Line: The debate hinges on consent and transparency. Does the government have to tell you when your Aadhaar data is used for AI decisions? Should you be able to opt out without losing access to services? India’s Supreme Court ruled in 2018 that Aadhaar can’t be mandatory for private services, but the law is still evolving—and AI is moving faster than the courts.

Final Thought: This isn’t just about India. The U.S. and EU are watching closely because India’s experiment with AI + digital IDs could become a model—or a cautionary tale—for the rest of the world.