Artificial Intelligence

Artificial Intelligence In the subject Technology, the topic Artificial Intelligence refers to the development of computer systems that can perform tasks requiring human intelligence. It includes areas like machine learning, speech recognition, and decision-making, making machines smarter and more efficient.

Artificial Intelligence

Definition of Artificial Intelligence (AI)

• “The Science and Engineering of making intelligent machines, especially intelligent Computer programs is Artificial intelligence” –JOHN MC CARTHY [Father of AI] 
• NITI Aayog defines AI as: AI is a constellation of technologies that enable machines to act with higher levels of intelligence and emulate the human capabilities of sense, comprehend and act. 
• Or simple ⇒ AI refers to the ability of machines to perform cognitive tasks like thinking, perceiving, learning, problem solving and decision making. 

History of Artificial Intelligence (AI)

• 1940s-1950s: Early Foundations
  • Alan Turing proposed the idea of machine intelligence and introduced the Turing Test to evaluate if machines can mimic human intelligence.
  • The first digital computers laid the groundwork for simulating human reasoning.
• 1956: Birth of AI as a Field
  • John McCarthy coined the term “Artificial Intelligence” during the Dartmouth Conference.
• 1960s: Early Progress
  • Development of early AI programs like ELIZA (a simple chatbot) and SHRDLU (language processing).
• 1970s: AI Winter Begins
  • Progress slowed due to limited computing power and lack of funding.
• 1980s: Revival of AI
  • Japan’s Fifth Generation Computer Project sparked renewed interest and funding in AI research.
  • Rise of Machine Learning: Introduction of neural networks and decision trees, enabling systems to improve over time.
• 1990s: Milestones in AI
  • Neural networks inspired by the structure of the human brain became more effective.
  • 1997: IBM’s Deep Blue defeated world chess champion Garry Kasparov, showcasing AI’s capability in strategic thinking.
• 2000s: Rapid Advancements
  • AI research accelerated with improved computing power (e.g., GPUs), and the rise of big data (e.g., ImageNet), Machine learning.
• 2010s-Present: AI in Everyday Life
  • Breakthroughs in deep learning revolutionized fields like image recognition, natural language processing, and autonomous vehicles → Siri, Alexa.

Types of Artificial Intelligence

• Based on Capabilities:
  • Narrow AI (Weak AI): Focused on performing a single task efficiently.
    • Example: Voice assistants like Siri, Alexa, and Google Assistant.
  • General AI (Strong AI): Hypothetical future AI capable of performing any intellectual task that a human can do.
    • Example: A robot capable of reasoning, problem-solving, and decision-making across multiple domains (yet to be achieved).
  • Superintelligent AI: AI surpassing human intelligence in all fields, including emotional intelligence.
    • Example: Theoretical concept; often depicted in sci-fi movies like Ex Machina.
• Based on Functionality:
  • Reactive Machines: Operate based on present data without storing past experiences.
    • Example: IBM’s Deep Blue chess computer.
  • Limited Memory: Can learn from past data to some extent.
    • Example: Self-driving cars analyzing traffic patterns.
  • Theory of Mind: Understand emotions and social interactions; still in experimental phases.
  • Self-Aware AI: Machines with consciousness and self-awareness (purely theoretical as of now)

Applications of AI

Sector	Applications	Examples
Healthcare	AI-driven diagnostic tools, personalized medicine, and medical research.	3Nethra: AI screening for eye diseases. AI-based drug discovery platforms.
Agriculture	AI in precision farming, commodity price forecasting, and autonomous farming robots.	Microsoft + KPAC: Agricultural commodity forecasting. E-Krishi Yantra: Multifunctional farming robot.
Industrial Revolution 4.0 & Manufacturing	AI-powered robots automate repetitive tasks, predictive maintenance, and product quality control.	Cobots in manufacturing. Predictive maintenance in factories. AI-driven quality control systems.
AI-Driven Personalization & Customer Service	AI personalizes products and services based on user preferences and behavior, and powers customer support systems like chatbots and virtual assistants.	Netflix: AI recommends content. SBI’s SIA: AI-powered customer chat assistant.
Cybersecurity & Fraud Detection	AI detects anomalous behavior to prevent cyberattacks, analyzes transaction patterns to prevent fraud, and strengthens security systems.	Darktrace Detect: AI-based cybersecurity for threat detection. AI in financial fraud detection.
Resource Management & Supply Chain	AI optimizes energy distribution, resource allocation, and logistics, balancing supply and demand in various sectors.	Smart Grids: AI for energy management. CoWIN app: AI-driven vaccine distribution. Telemedicine.
Digital Governance & Policing	AI streamlines governance functions, aids law enforcement, and enhances public service delivery.	CCTNS: AI in policing. SUPACE: AI for assisting judges in legal processes.
Disaster Management & Humanitarian Aid	AI predicts natural disasters, improves resource distribution, and aids in search and rescue operations.	Google AI: Predicts floods 7 days in advance. AI-powered search and rescue operations.
Space Research & Defense	AI enhances space missions and defense strategies, from real-time mission data analysis to autonomous drones and cyber defense.	Virtual Launch Control Centre (VSSC): AI for Chandrayaan-3 launch management. AI in cyber defense, autonomous drones.
Education	AI provides personalized learning experiences, automates administrative tasks, and enhances student outcomes.	BYJU’s: AI-powered tutoring system. Automated grading systems in schools.
Finance & Banking	AI enhances credit scoring, fraud detection, financial analysis, and high-frequency trading.	ICICI Bank: AI software robots. AI in credit scoring. Investment management platforms.
Retail & Hospitality	AI powers recommendation systems, dynamic pricing, and improves customer experience.	Amazon: AI-powered product recommendations. AI for hotel booking and dynamic pricing.
Transportation	AI improves safety and efficiency in transportation through self-driving cars and traffic management.	Tesla: Self-driving vehicles. AI for traffic management.
Environmental Protection	AI monitors climate, predicts weather changes, and aids in wildlife conservation.	AI in climate change modeling. AI in wildlife conservation and anti-poaching.
Media & Entertainment	AI automates content creation, video editing, and personalized user experiences.	AI-generated content. AI-based video editing tools.

Generative AI

• It is a cutting-edge technological advancement that utilizes machine learning and artificial intelligence to create new forms of media, such as text, audio, video, and animation.
• Generative AI models are trained using unsupervised learning techniques. They generate data by observing and capturing patterns and statistics from training data, creating new content that mimics human-generated data.
• The most used models are Variational Autoencoders (VAEs), Generative Adversarial Networks (GANs), and autoregressive models.
• Popular Generative AI Tools: ChatGPT (OpenAI), Gemini (google)

Challenges of AI

Challenges	Description	Examples
Bias and Fairness	AI systems can inherit biases from training data, leading to discriminatory or unfair outcomes.	MIT Media Lab study: Facial recognition systems showed higher error rates for darker-skinned individuals. Pulse oximeters are less accurate for individuals with darker skin tones.
Privacy Concerns	AI relies on vast amounts of personal data, raising concerns about privacy infringement and data misuse.	Cambridge Analytica scandal: Social media data used for voter manipulation during elections.
Algorithmic Transparency & Accountability	Difficulty in ensuring that AI systems align with ethical standards or legal requirements, especially when their decisions are opaque.	COMPAS algorithm: Used in the U.S. criminal justice system, found to disproportionately label Black defendants as higher risk.
Automated Cyber Attacks	Cybercriminals and State actors can leverage AI to enhance the efficiency and scale of cyberattacks.	AI-driven phishing campaigns, malware deployment, and DDoS attacks.
Unemployment & Job Displacement	Automation through AI could lead to the loss of jobs, potentially resulting in economic disruption.	McKinsey Global Institute prediction: AI will displace 85 million jobs by 2025.
Weaponization of AI	AI-driven technology is increasingly used for military or espionage purposes, potentially creating new global security risks.	Stuxnet worm: A cyber weapon developed by the U.S. and Israel to disrupt Iran’s nuclear enrichment facilities.
Disproportionate Power and Control	Major tech companies control large datasets and AI technology, creating an imbalance of power.	Data-oligarchic society: Technology giants dominate the AI landscape, stifling smaller competitors.
Social Engineering	AI techniques such as targeted advertising and personalized content recommendation exploit cognitive biases and manipulate behavior.	Social media: Algorithms prioritize sensationalist or polarizing content to reinforce existing biases.
Environmental Impact	The massive data centers required for AI processing contribute significantly to global carbon emissions.	Power-hungry air conditioning systems in data centers create a substantial carbon footprint.
Exacerbating Inequalities	Digital Divide: Unequal access to digital technologies and AI education.	Growing wealth disparity due to AI control by a few tech giants.
Legal and Regulatory Challenges	Absence of Comprehensive AI Laws: India lacks specific laws to regulate AI use, leaving a legal vacuum for issues like data protection and algorithmic accountability.Cross-Border Data Sharing: AI systems often process data stored in different countries, leading to jurisdictional conflicts.Intellectual Property Rights (IPR) Issues: Uncertainty around the ownership of AI-generated works—does the credit go to the machine or the programmer?AI in Criminal Justice: Use of AI in predictive policing raises questions about profiling and presumption of guilt without evidence.
Deep Fake	Deep Fakes are digitally synthesized media created using advanced deep learning techniques like Generative Adversarial Networks (GAN), to produce realistic but entirely fabricated visual and audio content. For example, a viral video featuring an actress in a swimsuit or the Prime Minister performing Garba. The incorporation of hyper-realistic digital falsification presents the following challenges:Misinformation and Disinformation: Spread false information, manipulate public opinion, damage credibility. (E.g., deepfake of Ukrainian President Zelenskyy)Impersonation and Identity Theft: Lead to financial fraud.Privacy Violations: Use personal photos or videos without consent, lead to privacy violations.Legal and Ethical Concerns of AI: Raise questions about legality and ethics, call for regulation and accountability.Weaponization Against Women: Primarily pornographic, causing psychological trauma and social repercussions.Threats to National Security: Undermine public safety, create chaos, stir anti-state sentiments.Undermine Trust in Democratic Institutions: Manipulate people, sabotage elections. (E.g., viral video of Manoj Tiwari)Declining Trust in Traditional Media: Causes short-term and long-term harm.Contribution to the “Liar’s Dividend”: Dismiss undesirable truths as fake news, undermine credibility.To ensure the safety of Digital Nagrik, MeitY recently sent social media platforms advisories on deepfakes under IT Rules 2021.

National Initiatives to Promote AI

Initiative/Program	Purpose/Description
National AI Portal (INDIAai Portal)	Joint initiative by the Ministry of Electronics and IT (MeitY), National e-Governance Division (NeGD) and NASSCOM.Launched in May 2020, serves as a national AI resource hub, offering educational materials, news, and updates on AI developments in India.
National Program on Artificial Intelligence (India)	Domestic initiative focusing on responsible AI, innovation, and balancing the benefits of AI with potential risks.
Future Skills Prime	Program promoting reskilling and upskilling in AI-related skills to prepare the workforce for the changing job landscape.
SUPACE (Supreme Court Portal for Assistance in Courts Efficiency)	AI-based portal to assist judges with legal research, improving efficiency in the judicial system.
Global Partnership on Artificial Intelligence (GPAI)	India became a part of the GPAI in June 2020, collaborating on global AI research and development.
US India Artificial Intelligence (USIAI) Initiative	Launched to scale up the science and technology relationship between India and the United States in AI.
National Mission on Interdisciplinary Cyber-Physical Systems	Technology Innovation Hubs (TIH)Establishment of AI and ML hubs at IIT Kharagpur for technology innovation and research in AI.
Artificial Intelligence Research, Analytics and Knowledge Assimilation Platform	A cloud computing platform aimed at making India a leader in AI and transforming sectors like education, health, agriculture, urbanization, and mobility.
Responsible Artificial Intelligence (AI) for Youth	A national program aimed at educating the youth about responsible AI practices.
India AI Mission	Rs 10,300 crore initiative to procure AI compute capacity and offer it at subsidized rates to start-ups and researchers.
BharatGen Initiative	Government-funded multimodal AI initiative.

BharatGen Initiative: Government-funded Multimodal AI Initiative

• Goal: Develop generative AI models in Indian languages, speech, and computer vision to boost public service delivery and citizen engagement.
• Vision: Develop inclusive, efficient AI technologies tailored to India’s socio-cultural and linguistic landscape.
• Spearheaded by: IIT Bombay under the National Mission on Interdisciplinary Cyber-Physical Systems (NM-ICPS) of the Department of Science & Technology (DST)
• Timeline: Expected completion in two years.

Key Features:

• Multilingual & Multimodal Models: Focused on India’s linguistic and cultural diversity.
• Bharat-centric Data: Training on India-specific data for accurate representation.
• Open-Source Platform: Ensuring accessible AI for all sectors.
• Strategic Impact: Reflects India’s commitment to ‘Make AI in India, Make AI for India’

IndiaAI Mission

The IndiaAI Mission was launched on March 7, 2024, with a budget outlay of ₹10,371.92 crore, to position India as a global AI leader and democratize AI benefits across society.

Objectives:

• Democratize access to AI technologies.
• Foster technological self-reliance and innovation.
• Promote ethical and responsible AI usage.
• Drive inclusive growth and socio-economic transformation.

Implementation Agency: IndiaAI, an Independent Business Division (IBD) under the Digital India Corporation (DIC), Ministry of Electronics and IT (MeitY).

7 Key Pillars of the Mission:

1. IndiaAI Compute Capacity
   • Build a high-end scalable AI computing ecosystem with 10,000+ GPUs through public-private partnerships.
   • Launch an AI marketplace offering AI as a service and pre-trained models.
2. IndiaAI Innovation Centre
   • Develop indigenous Large Multimodal Models (LMMs) and sector-specific foundational models.
3. IndiaAI Datasets Platform
   • Create a unified platform for seamless access to high-quality, non-personal datasets for startups and researchers.
4. IndiaAI Application Development Initiative
   • Promote AI solutions in critical sectors through collaboration with Central Ministries, State Departments, and institutions.
5. IndiaAI FutureSkills
   • Increase AI education in undergraduate, master’s, and Ph.D. programs.
   • Establish Data and AI Labs in Tier 2 and Tier 3 cities for foundational AI training.
6. IndiaAI Startup Financing
   • Support deep-tech AI startups with streamlined funding access for innovative AI projects.
7. Safe & Trusted AI
   • Develop frameworks, tools, and self-assessment checklists for responsible AI development.

Expected Outcomes:

• Strengthen India’s tech sovereignty and global AI competitiveness.
• Generate highly skilled employment opportunities to leverage India’s demographic dividend.
• Catalyze AI-led socio-economic transformation through impactful, scalable solutions.
• Establish India as a global leader in ethical AI innovation for social good.

The IndiaAI Mission exemplifies India’s commitment to harnessing AI’s transformative potential for inclusive development and global leadership.

Steps taken

AI in Governance

• BHASHINI: An AI-based language translation tool to support Indian languages.
• India Stack and AI: Digital infrastructure that facilitates AI-driven services.

AI Computer & Systems

• AIRAWAT: India’s AI supercomputer at C-DAC, Pune.
• National Supercomputing Mission: Aimed at strengthening India’s computing capabilities.
• MeitY Quantum Computing Applications Lab: Focuses on developing quantum computing applications.

Data for AI

• Data Management Office: Ensures the standardization of data management practices.
• India Datasets Program & India Data Platform: Provides access to non-personal datasets for startups and researchers.

AI, Intellectual Property (IP) & Innovation

• Centre of Excellence for Artificial Intelligence: Promotes AI research and development.
• MeitY Start-up Hub: Supports AI-based start-ups.
• Proposed National Centre on AI (NCAI): Plans for a centralized AI research hub.

Skilling in AI

• Future Skills Prime: A joint initiative by Nasscom & MeitY offering AI certification programs.
• Responsible AI for Youth: A program aimed at making school students AI-ready.

AI Ethics & Governance

• RAISE: Responsible AI for Social Empowerment, by MeitY, focusing on ethical AI use.
• Digital Sansad App: An AI-powered platform for transcribing proceedings in the new Parliament

AI Regulation

Why AI Needs Regulation:

1. Managing Unpredictable Risks
2. The “Black Box” Challenge:
   • Some AI systems are so complex that they function like a “black box”—even the developers who create them don’t fully understand how they arrive at decisions.
3.  Tackling Inaccuracies and Bias
4. Uncertainty in Behavior
5. National Security and Safety Concerns.
6. Danger of AI being controlled by a few multinational corporations.
7. Cyber Risks in the Age of Generative AI:
   • AI-Driven Phishing: Phishing emails increased by 1,265% since 2022. AI creates convincing emails that are hard to detect.

Global AI Governance Approaches

• 2018: Canada and France announced plans for a G7-backed International Panel on AI, modeled after the IPCC.
• 2019: Panel renamed as Global Partnership on AI (GPAI).

GPAI Launch (June 2020):

• 15 founding members: Australia, Canada, EU, France, Germany, India, Italy, Japan, S. Korea, Mexico, New Zealand, Singapore, Slovenia, UK, USA.
• 2023: GPAI grew to 29 members.
• Mandate: Focus on responsible AI, data governance, future of work, and innovation/commercialization.
• GPAI is hosted by the Organisation for Economic Co-operation and Development (OECD).
• India is the lead chair of GPAI in 2024

International AI Regulations & Principles:

• 2019: OECD AI Principles & G20 AI Principles.
• 2020: European Union drafts AI regulation strategy.
• 2021: UNESCO adopted a global instrument on the ethics of AI.
• 2023: Hiroshima AI Process (HAP) by G7.
• 2023: AI Safety Summit, held at Bletchley Park in the UK from November 1–2, 2023
• 2024: U.N. AI Advisory Body’s Final Report – 7 Recommendations 
• 2024: First Global Legally Binding Treaty on AI → Council of Europe Framework Convention on Artificial Intelligence, Human Rights, Democracy, and the Rule of Law

AI for Good (ITU & UN):

• Platform to leverage AI for UN Sustainable Development Goals.

Different Approaches to Regulate AI 

1. European Union (EU): Adopted a comprehensive AI Act to regulate AI use, focusing on transparency and risk management.
   • Risk-Based Classification of AI Applications
2. United States: Advocates for voluntary compliance instead of strict laws, promoting industry self-regulation.
3. China: Focuses on state control and maintaining social stability in AI development.

India’s Approach to Regulate AI

• Sovereign AI : India is taking a multifaceted approach to regulate AI, focusing on its sovereign applications in critical sectors like healthcare and agriculture while ensuring privacy and ethical considerations are addressed. 
• Digital Personal Data Protection Act in 2023: address concerns related to AI platforms by regulating how personal data is handled, ensuring individuals’ privacy is protected.
• Global Partnership on Artificial Intelligence: As a member of GPAI, India hosted the 2023 GPAI Summit in New Delhi, focusing on responsible AI, data governance, and the future of work, innovation, and commercialization.
• The National Strategy for Artificial Intelligence #AIForAll strategy, by NITI Aayog: This strategy outlines AI R&D guidelines focusing on healthcare, agriculture, education, smart cities and infrastructure, and smart mobility and transformation
• EAC-PM Paper on AI Regulation: This paper proposes a novel AI regulation framework based on a Complex Adaptive System with five key principles:
  • Specialist Regulator: Establishing an expert regulatory body with a broad mandate and swift responsiveness.
  • Transparency: Ensuring open licensing of core algorithms for external audits and continuous monitoring.
  • AI Accountability: Mandating standardized incident reporting protocols and defining clear lines of accountability.
  • Manual Overrides and Authorization Chokepoints: Empowering human intervention in AI system operations.
  • Guardrails and Partitions: Implementing boundary conditions to limit undesirable AI behaviors.
• The MEITY  issued an advisory to several large platforms for the regulation of generative Artificial Intelligence (AI) recently. MeitY stipulated that they must seek government permission to operate in India, provide disclaimers and disclosures during testing phases, and ensure their resources do not permit bias, discrimination, or threats to electoral integrity through AI.

Thus, India’s approach emphasizes a balanced path, fostering AI innovation while ensuring ethical use and safeguarding individual rights.

GPAI Summit 2023

India, as the Chair of the Global Partnership on Artificial Intelligence (GPAI), hosted the GPAI Summit from December 12-14, 2023, at Bharat Mandapam, New Delhi. 

Major Outcomes

• Adoption of the GPAI New Delhi Declaration, emphasizing safe, secure, and trustworthy AI and sustainability of GPAI projects.
• Prime Minister’s call for a global framework for ethical AI use.
• Endorses India’s proposal to establish and maintain a Global Digital Public Infrastructure Repository (GDPIR) to share digital public goods.
• India highlighted as a key player in AI talent and innovation and a global hub for AI development.
• Emphasis on AIRAWAT (AI Research Analytics and Knowledge Dissemination Platform) and the National Program on Artificial Intelligence.
• YUVAi Initiative: A program launched to train school students (8th to 12th grade) in AI, empowering them to use AI for solving societal issues and contributing to inclusive development.

Key Highlights of PM Modi’s Address:

Government’s Approach to AI

• Guided by the principle of ‘AI for All’ to ensure social development and inclusive growth.
• Launch of a National Program on Artificial Intelligence and the upcoming AI Mission to enhance AI computing capabilities and promote startups.
• Extension of AI-related skills to Tier 2 and Tier 3 cities through training institutes.

Ethical and Responsible AI

• Emphasis on responsible human-centric AI governance with focus on ethics, transparency, and accountability.
• Proposal for a global framework for ethical AI use, following the G20 New Delhi Declaration on AI Principles.
• Suggestions for AI Regulation:
  • Explore software watermarking to label AI-generated content.
  • Audit mechanism: Categorize AI tools (red, yellow, green) based on capabilities.

AI’s Role in Employment and Education

• Standardized AI education curriculum and resilient employment measures.
• Incorporating upskilling and reskilling into the AI growth curve to prepare the workforce.

Inclusivity and Democratic Values in AI Development

• AI should promote equality and social justice, avoiding the unequal access to technology seen in the past.
• Use of AI to enhance digital inclusion by supporting local languages and reviving extinct languages.
• Leveraging AI to advance knowledge in fields like Vedic mathematics and Sanskrit literature.

AI as a Double-Edged Sword

• While transformative, AI can be potentially destructive (e.g., deepfakes, cybersecurity threats).
• The need for countermeasures to prevent misuse by malicious actors.

India’s Vision for AI

• AI as a tool for sculpting a transformative and sustainable future.
• Collaboration to ensure AI’s development adheres to human and democratic values.
• Promoting trust in AI by addressing its ethical, economic, and social implications.

U.N. AI Advisory Body’s Final Report: Seven Key Recommendations for Global AI Governance

1. Establish a Scientific Panel on AI to provide reliable, scientific knowledge about AI. It will Address the information asymmetry between AI developers and the global community.
2. Policy Dialogue on AI Governance
3. Establish an AI Standards Exchange platform to promote global standards for AI systems, ensuring consistency and security.
4. Global AI Capacity Development Network
5. Global AI Fund
6. Global AI Data Framework for data transparency and accountability in AI systems to ensure ethical usage.
7. AI Office to support and coordinate the implementation of these recommendations.

Council of Europe Framework Convention on Artificial Intelligence, Human Rights, Democracy, and the Rule of Law 

• It establishes a legally binding framework to ensure that AI systems align with human rights, uphold democratic values, and operate within the bounds of the rule of law.

What the Treaty Commits Nations To

1. Protection of Human Rights
   • Member states must ensure that AI systems respect privacy, equality, and other human rights.
2. Safeguarding Democratic Institutions: AI cannot undermine democratic processes or free speech.
3. Accountability for AI Outcomes: Signatories will be responsible for preventing harmful or discriminatory outcomes from AI systems, whether developed in the public or private sectors.
4. Applicability Across Geographies and Sectors:The treaty applies to AI systems worldwide, with exceptions for national security and research.

Significance of the Treaty

1. First Global AI Pact
2. Global Participation
3. Precedent for Global AI Governance

Challenges and Concerns

1. Lack of Enforcement Mechanisms.
2. Limited Scope.