Lecture Notes: Introduction to Artificial Intelligence

1. What is Artificial Intelligence?

Artificial Intelligence (AI) is the field of computer science dedicated to creating systems that can
perform tasks that normally require human intelligence.
These tasks include:

 Problem-solving

 Learning from experience

 Understanding natural language

 Recognizing patterns

 Decision-making

2. Core Definitions

 Weak AI (Narrow AI): Systems designed to perform specific tasks (e.g., chatbots,
recommendation systems).

 Strong AI (General AI): Hypothetical AI that can perform any intellectual task a human can.

 Machine Learning (ML): A subset of AI that allows systems to learn from data.

 Deep Learning: A type of ML that uses neural networks with many layers.

3. Brief History of AI

3.1 Early Foundations

 1940s–1950s: Theoretical groundwork by Alan Turing ("Can machines think?").

 Development of early digital computers.

3.2 The Birth of AI

 1956: Dartmouth Conference — the term "Artificial Intelligence" coined by John McCarthy.

 Early programs solved logic problems and played games.

3.3 The AI Winters

 1970s & late 1980s: Funding and interest dropped due to limited progress.

3.4 Modern AI Boom

 2000s onwards: Advances in computational power, big data, and algorithms led to
breakthroughs in speech recognition, vision, and language processing.
4. Key AI Techniques

4.1 Search and Optimization

 Algorithms like A*, Dijkstra’s used for pathfinding and problem-solving.

4.2 Knowledge Representation

 Ontologies and logic-based systems to store and use knowledge.

4.3 Machine Learning

 Supervised learning: Model trained on labeled data.

 Unsupervised learning: Model finds hidden patterns in data.

 Reinforcement learning: Model learns by trial and error, receiving rewards or penalties.

4.4 Natural Language Processing (NLP)

 Understanding and generating human language.

 Applications: translation, sentiment analysis, chatbots.

4.5 Computer Vision

 Interpreting images and videos.

 Applications: facial recognition, medical imaging.

5. AI System Components

 Data: Raw material for training and improving models.

 Algorithms: Mathematical instructions for learning and decision-making.

 Computing Power: Hardware like GPUs and TPUs.

 Evaluation Metrics: Accuracy, precision, recall, F1 score.

6. Applications of AI

6.1 Healthcare

 Diagnosis assistance (e.g., detecting cancer in scans).

 Drug discovery.

 Personalized treatment recommendations.

6.2 Finance

 Fraud detection.

 Algorithmic trading.

 Customer service chatbots.

 Self-driving cars.

 Traffic management systems.

6.4 Education

 Adaptive learning platforms.

 Automated grading.

6.5 Manufacturing

 Predictive maintenance.

 Quality control using computer vision.

6.6 Everyday Life

 Voice assistants (Siri, Alexa).

 Personalized recommendations (Netflix, YouTube).

7. Benefits of AI

 Handles repetitive and dangerous tasks.

 Processes vast amounts of data quickly.

 Can work continuously without fatigue.

 Improves decision-making with predictive analytics.

8. Challenges and Limitations

 Bias in AI: Models can reflect and amplify human biases.

 Transparency: “Black box” problem where decisions are hard to explain.

 Data Requirements: AI often needs large, high-quality datasets.

 Ethical Concerns: Privacy, surveillance, and job displacement.

9. Ethics in AI

9.1 Responsible AI Principles

 Fairness: Avoid discrimination.

 Transparency: Explain how decisions are made.

 Accountability: Assign responsibility for outcomes.

 Privacy: Protect user data.

 Use of AI in autonomous weapons.

 Deepfake technology.

 AI in law enforcement.

10. The Future of AI

 General AI: Moving toward human-level intelligence remains uncertain.

 AI and Sustainability: Optimizing energy, reducing waste.

 Human-AI Collaboration: Augmenting human abilities rather than replacing them.

11. Summary Table

Concept Description Example Application

Weak AI Task-specific AI Voice assistants

Machine Learning Learning from data Spam email detection

Deep Learning Neural networks with many layers Image recognition

NLP Understanding human language Chat translation

Computer Vision Interpreting images Self-driving cars

12. Discussion Questions

1. What are the main differences between narrow AI and general AI?

2. How can bias enter into AI systems, and how can it be mitigated?

3. Should AI systems be given legal accountability?

13. Conclusion

Artificial Intelligence is reshaping industries, economies, and daily life. While the technology offers
unprecedented opportunities for efficiency, creativity, and problem-solving, it also brings complex
challenges related to ethics, fairness, and social impact. The future of AI depends not only on
technical breakthroughs but also on responsible development, global cooperation, and continuous
dialogue between technologists, policymakers, and the public.