Maarg Manthan · Topic 15.10

Artificial Intelligence: Concepts and Applications

Science & Technology › Materials Science, Nanotechnology and Robotics · Topic 15.10

Artificial intelligence is the ability of a machine to perform tasks that normally need human intelligence, such as learning, reasoning, understanding language, seeing and making decisions. The field began at the Dartmouth workshop of 1956, where John McCarthy coined the term, and Alan Turing had proposed the imitation game test in 1950. Machine learning lets systems learn from data instead of fixed rules, deep learning uses many layered neural networks, and generative AI creates text, images, code and audio, as large language models do. Narrow AI does one task, general AI would match human ability across tasks, and superintelligence is a speculative idea. AI is used in healthcare, agriculture, finance, education, governance, translation and defence. India approved the IndiaAI Mission in March 2024.

5 min read · UPSC Prelims · Science & Technology · Materials Science, Nanotechnology and Robotics

At a Glance

Term coined McCarthy, 1956
Test Turing, 1950
Subfields ML, deep learning, NLP
Generative AI Creates new content
Indian mission IndiaAI, March 2024
Exam link UPSC Prelims, Science & Technology: Materials Science, Nanotechnology and Robotics

Where Does It Fit?

This topic follows the account of robotics applications. Artificial intelligence provides the sensing and decision making for modern robots and links with earlier topics on cloud, big data and quantum computing. Previous topic: Applications of Robotics: Industry, Aerospace, Healthcare, Military, Agriculture, Disaster Response. Next topic: Advantages and Disadvantages of Artificial Intelligence and Robotics.

What Are the Key Concepts and History of AI?

Concept	Meaning
Turing test	Proposed by Alan Turing in 1950: a machine passes if a human judge cannot tell its answers from a human’s
Dartmouth, 1956	The summer workshop where John McCarthy introduced the term artificial intelligence, and the field began
Machine learning	Algorithms that improve with data. Supervised learning uses labelled data, unsupervised finds patterns, and reinforcement learning learns by reward
Neural networks and deep learning	Layers of artificial neurons trained on large data. Deep learning drove advances in vision and speech after 2012
Natural language processing	Lets machines read, write and translate language
Generative AI	Models that produce new text, images, audio, video or code. Large language models predict the next word after training on huge text data
Narrow, general and super	Narrow AI does specific tasks and is all that exists today. Artificial general intelligence would be human level across tasks

What Are the Applications and Indian Initiatives?

Healthcare: reading X-rays and retinal scans, drug discovery and triage
Agriculture: crop advice, pest detection and price forecasts, such as AI based advisory chatbots for farmers
Governance: fraud detection, traffic management, translation and grievance handling
Finance: credit scoring, fraud alerts and chatbots
Education: personalised learning and language tools
Climate and disaster: AI weather models and flood forecasting
National Strategy for AI, 2018: NITI Aayog’s paper titled AI for All, focusing on healthcare, agriculture, education, smart cities and mobility
IndiaAI Mission: approved by the Union Cabinet in March 2024 with an outlay of about Rs 10,372 crore, for computing capacity, datasets, foundation models, skills and safe AI

What Else Should You Remember?

Bhashini: National Language Translation Mission, started in 2022, to give translation and speech tools in Indian languages
Responsible AI for Youth: a programme for school students launched in 2020
AI Kosh: a platform of datasets under the IndiaAI Mission
Compute capacity: the Mission plans a pool of tens of thousands of GPUs for researchers and start-ups
Foundation models: support for Indian language models
Global Partnership on AI: India is a founding member and chaired it in 2024
AI summits: Bletchley Park in 2023, Seoul in 2024 and Paris in 2025 on AI safety and action; India was to host an AI Impact Summit in 2026, and the details should be checked
Algorithmic bias: AI can copy bias in training data and treat groups unfairly
Explainability: deep networks are often black boxes whose decisions are hard to explain
Deepfakes: AI made fake media, as in Contemporary Debates: Big Data and Privacy, Cybercrime and Security
Hallucination: generative models can state false things confidently
Data and privacy: training needs large data, which raises consent questions under the data protection law
European Union AI Act: the first comprehensive law, with risk based rules, entered into force in 2024
Compute and cloud: AI training needs large clusters, as in Cloud Computing and Big Data
Quantum link: quantum methods may later help AI, as in Quantum Computing
Chip dependence: advanced AI chips are made by few firms, which makes supply chains strategic
Military AI: autonomous weapons raise ethical and legal questions

Why Does It Matter for the Exam?

Terms: ML, deep learning, generative AI
History: Turing 1950, Dartmouth 1956
India: NITI strategy 2018, IndiaAI Mission 2024, Bhashini
Current affairs link: AI summits and regulation
Common errors: confusing machine learning with all of AI or saying general AI exists

As an analytical point, AI is a general purpose technology that can improve services at scale, but it needs data, compute and skills. Safe use requires rules on bias, privacy and accountability along with investment in Indian language models.

Why Is This Topic Important?

Concepts ML to generative AI
Uses Many sectors
India Mission and Bhashini

What Came Next?

The next topic covers the advantages and disadvantages of artificial intelligence and robotics. Next topic: Advantages and Disadvantages of Artificial Intelligence and Robotics.

Key People and Terms

AI Machine intelligence
ML Learning from data
Deep learning Multi-layer neural networks
LLM Large language model
NLP Natural language processing
Bias Unfair skew in output

Exam Corner

Points to Remember

Turing test: 1950; Dartmouth: 1956.
NITI strategy: 2018, AI for All.
IndiaAI Mission: March 2024, about Rs 10,372 crore.
Bhashini: 2022.
EU AI Act in force: 2024.

Do Not Confuse With

AI and machine learning Whole field versus learning methods
Narrow and general AI One task versus any task
Generative and discriminative Creates content versus classifies

Memory Hook

Turing asked, McCarthy named, data teaches, models generate; India missions in twenty-four.

Mains Angle

Questions ask for terms, history and Indian initiatives. These points help.

Concepts Types and methods
Uses Sectors
Governance Risks and rules

A Question You May Face

An original practice question, not a past paper question.

“Discuss the potential of artificial intelligence for India’s development and the regulatory challenges it poses.”

How to Answer

Concept ML to generative.
Uses Sectors.
Policy IndiaAI Mission.
Risks Bias, privacy.

GS Relevance

Prelims: AI terms, history, IndiaAI Mission, Bhashini. GS Paper 3: science and technology, IT, governance applications.

Frequently Asked Questions

What is artificial intelligence?

Artificial intelligence is the ability of computer systems to perform tasks that normally require human intelligence, such as learning from experience, understanding language, recognising images, reasoning and making decisions. Most systems today are narrow AI, built for specific tasks like translation or image recognition.

What is the difference between AI, machine learning and deep learning?

Artificial intelligence is the broad field of making machines intelligent. Machine learning is a part of it in which systems learn patterns from data instead of following fixed rules. Deep learning is a part of machine learning that uses many layered neural networks and works well for images, speech and language.

What is generative AI?

Generative AI is a kind of artificial intelligence that creates new content, such as text, images, audio, video or computer code, after learning from large amounts of data. Large language models, which predict the next word in a sentence, are an example. They can also make false statements, called hallucinations.

What is the IndiaAI Mission?

The IndiaAI Mission, approved by the Union Cabinet in March 2024 with an outlay of about Rs 10,372 crore, aims to build a national AI computing capacity, datasets platform, Indian foundation models, skills, start-up support and safe and trusted AI. It is run through the Ministry of Electronics and Information Technology.

What is the Turing test?

The Turing test, proposed by Alan Turing in 1950, asks whether a machine can hold a conversation so well that a human judge cannot reliably tell it apart from a person. It is a famous idea about machine intelligence, though many experts feel it does not measure real understanding.

PYQ Practice — Statement Analysis

1 The term artificial intelligence was coined in 1956.

True

By John McCarthy at Dartmouth.

2 Deep learning is the broad field that includes all of AI.

False

It is a part of machine learning.

3 The IndiaAI Mission was approved in 2024.