Here are three essay questions with detailed answers, each designed to be

worth approximately 10 marks, based on the content of the provided AI course

overview video.

Essay Questions and Answers for AI Course Overview

Question 1: Discuss the various perspectives on Artificial Intelligence
as presented in the video. How do these different viewpoints
contribute to a comprehensive understanding of AI?
Answer (10 Marks):
The video elucidates several key perspectives on Artificial Intelligence,
highlighting the multifaceted nature of this field. A comprehensive
understanding of AI requires considering these different viewpoints, as each
emphasizes a distinct aspect of what it means for a machine to be "intelligent."

Firstly, AI can be viewed as the behavior of machines that would be

considered intelligent if performed by humans.1 This perspective focuses
on the observable outcomes and functionalities of AI systems. For instance, if a
computer can play chess at a grandmaster level, it is deemed intelligent
because such a feat requires high intelligence in a human. This viewpoint is
pragmatic and performance-oriented, assessing AI based on its ability to mimic
or surpass human capabilities in specific tasks.
Secondly, AI is seen as making computers do things that people
currently do better.2 This perspective emphasizes the ongoing challenge
and frontier of AI research. It acknowledges areas where human cognitive
abilities still outperform machines, such as common sense reasoning,
creativity, or nuanced social interaction.3 This drives research towards
automating increasingly complex and human-centric tasks, pushing the
boundaries of what computers can achieve.
A third perspective defines AI as simulating the workings of the human
mind. This approach delves into the cognitive science aspect of AI, aiming to
understand and replicate the internal processes, thought patterns, and
decision-making mechanisms of the human brain. This perspective is more
concerned with the how of intelligence rather than just the what, potentially
leading to AI systems that learn and reason in a human-like fashion.
Alan Turing's view offers a behavioral test: AI actions are indistinguishable
from humans. This is famously embodied in the Turing Test, where a human
interrogator cannot tell if they are communicating with a machine or another
human.4 This perspective sets a high bar, requiring AI to master natural
language, general knowledge, automated reasoning, and machine learning to
exhibit truly human-like conversational abilities. It moves beyond task-specific
intelligence to a more holistic measure of human-level performance.
Finally, AI is described as the intelligent connection of perception to
action. This view emphasizes the practical application of AI in real-world
scenarios, particularly in fields like robotics. It highlights the crucial link
between sensing information from the environment (perception) and
generating appropriate responses or behaviors (action). This perspective is
vital for developing autonomous systems that can operate effectively in
dynamic and unpredictable environments.
Together, these perspectives provide a robust framework for understanding AI.
They cover the spectrum from performance-based evaluation to cognitive
replication, from pushing current human boundaries to achieving human-level
indistinguishability, and from abstract reasoning to practical, real-world
interaction. This multi-faceted approach allows for diverse research avenues
and a comprehensive assessment of AI's progress and potential.

Question 2: Explain the essential abilities required for an Artificial

Intelligence system to be considered truly intelligent. Provide
examples of how these abilities manifest in practical AI applications.
Answer (10 Marks):
For an Artificial Intelligence system to be considered truly intelligent, it needs
to possess a range of essential abilities that mirror, and often enhance, human
cognitive functions. The video highlights several of these crucial capabilities.

Firstly, a core ability is learning new concepts and tasks. This is

fundamental for AI to adapt and improve without explicit reprogramming for
every new situation. For instance, in deep learning, a system can learn to
identify new objects in images after being exposed to a vast dataset, rather
than being programmed with individual rules for each object.5 This ability
underpins advancements in machine learning and deep learning, enabling AI to
continuously refine its performance.6
Secondly, reasoning and drawing conclusions is vital. Intelligent systems
should be able to process information, apply logical rules, and infer new
knowledge. Expert systems, for example, use a knowledge base and inference
engine to diagnose medical conditions or provide financial advice by reasoning
through symptoms or financial data.7 This capability allows AI to solve
problems that require logical thought and deduction.8
Thirdly, the ability to remember and draw conclusions from complex
facts is essential, especially for systems dealing with large amounts of
information. This is closely related to knowledge representation, where
information is stored and retrieved efficiently. In legal AI, systems can sift
through vast legal precedents and statutes, remembering relevant cases and
drawing conclusions to assist lawyers, a task that would be overwhelming for
humans alone.9
Furthermore, understanding natural language and perceiving visual
scenes are critical for AI to interact with the human world.10 Natural Language
Processing (NLP) enables virtual assistants like Siri or Google Assistant to
comprehend spoken commands and respond meaningfully.11 Computer vision
allows self-driving cars to perceive their surroundings, identify pedestrians,
traffic signs, and other vehicles, making real-time decisions based on visual
input.12
Moving around (robotics) and planning sequences of actions are
intertwined abilities necessary for physical AI systems.13 Robots performing
tasks in factories or exploring remote planets need to navigate their
environment, avoid obstacles, and execute a series of steps to achieve a
goal.14 Robotic arms in manufacturing plants plan precise sequences of
movements to assemble products efficiently, demonstrating highly coordinated
action planning.15
Finally, an intelligent AI system should be capable of offering advice based
on rules and situations and performing tasks requiring high
intelligence. This extends beyond simple computation to complex problem-
solving. Beyond expert systems, AI in financial trading can analyze market
data, identify patterns, and offer investment advice.16 In scientific discovery,
AI can analyze experimental results, formulate hypotheses, and suggest new
experiments, performing tasks that require significant intellectual capacity.17
In essence, these abilities collectively enable AI systems to perceive, process,
reason, learn, and act intelligently in diverse environments, pushing them
beyond mere automation towards true cognitive emulation.

Question 3: AI is described as a multidisciplinary field. Discuss the

significant contributions from at least five different parent disciplines
that have shaped and continue to influence Artificial Intelligence.
Answer (10 Marks):
Artificial Intelligence is inherently multidisciplinary, drawing foundational
knowledge and methodologies from a diverse array of academic fields.18 This
cross-pollination has been crucial in shaping AI into the complex and rapidly
evolving domain it is today. The video highlights several key parent disciplines:
Firstly, Computer Engineering/Science forms the bedrock of AI. This
discipline provides the hardware infrastructure, computational models,
algorithms, and programming languages necessary for AI systems to exist and
function.19 Concepts like data structures, algorithms, parallel processing, and
distributed systems are fundamental.20 More recently, advancements in data
science, machine learning, and deep learning algorithms, which are core to
modern AI, originate directly from computer science. Without robust
computational capabilities and efficient programming paradigms, AI would
remain a theoretical concept.
Secondly, Philosophy has profoundly influenced AI, particularly in establishing
its conceptual and logical foundations.21 Philosophers have long debated the
nature of knowledge, reasoning, consciousness, and the mind-body problem.22
These discussions laid the groundwork for AI concepts such as logic (e.g.,
propositional, predicate logic), knowledge representation, and the very idea of
machine intelligence. Questions about what constitutes intelligence and how
knowledge is acquired and used in reasoning were first explored in philosophy,
providing the theoretical underpinnings for AI research.
Thirdly, Mathematics provides the formal language and tools for AI.23 From
logic and set theory to probability and statistics, mathematics is indispensable
for representing information, proving theorems, developing algorithms, and
handling uncertainty.24 For example, machine learning algorithms heavily rely
on linear algebra, calculus, and probability theory for optimization, pattern
recognition, and statistical inference.25 The ability to mathematically model
problems and derive solutions is central to almost every AI technique.
Fourthly, Psychology and Cognitive Science offer crucial insights into
human intelligence, perception, and learning.26 AI aims to build intelligent
systems, and understanding how humans perceive, learn, remember, and solve
problems provides valuable inspiration and models.27 Concepts like knowledge
representation, abstraction, and human problem-solving strategies from
cognitive psychology directly inform AI research in areas like expert systems,
cognitive architectures, and even neural networks (inspired by biological
brains).
Finally, Linguistics is a critical parent discipline, especially for areas like
Natural Language Processing (NLP).28 Linguistics provides the theories and
models for understanding the structure, meaning, and use of human language.
This includes syntax, semantics, phonetics, and pragmatics. AI systems that
can understand, generate, and translate human language (e.g., chatbots,
machine translation) are built upon decades of linguistic research, enabling
them to process and interact with the complexities of human
communication.29
In conclusion, AI's multidisciplinary nature means that progress in one field
often catalyzes breakthroughs in another. The synergistic contributions from
computer science, philosophy, mathematics, psychology, and linguistics,
among others, ensure that AI continues to evolve as a field that is both
technologically advanced and deeply rooted in fundamental principles of
intelligence and cognition.