Unit 15 Fundamentals of Artificial Intelligence (AI) & Intelligent Systems (K/618/5660) Assignment Brief 2026

Unit 15 Fundamentals of Artificial Intelligence (AI) & Intelligent Systems Assignment Brief 2026

This unit is aligned to the Microsoft Azure AI Fundamentals Certification. See section 3.1.4 for further guidance on claiming certification.

Introduction

Intelligent Systems are revolutionising industry and changing the way we accomplish daily routines. They help to introduce flexibility, quality and energy efficiency to name a few to an increasing range of applications. For example, transportation, healthcare, education, and the defence sector. Intelligent Systems are enabled by various underpinning technologies, especially Artificial Intelligence (AI). AI offers opportunities to gain insights from data or perceive the environment to take intelligent actions that maximize the chances of performing a task faster or not previously possible.  The growth in AI potential offers companies opportunities to reduce costs, increase productivity and introduce new products to the market. Therefore, people skilled in AI and its applications are in high demand.

This unit is designed to introduce the science behind machine intelligence and the philosophical debate around the ambitions of simulating human intelligence to solve real-world problems. Students will be guided to appreciate AI types and applications and develop a better understanding of aspects related to intelligent agents.  Other topics included in the unit covers Artificial Narrow Intelligence (ANI), Artificial General Intelligence (AGI), Ambient Intelligence, the major differences between top-down and bottom-up approaches to AI, and understanding Machine Learning (ML) algorithms (e.g. SVM, Naïve Bayes, Random Forest and KNN) and processes including dataset preparation.

On successful completion of this unit, students will be able to investigate AI fundamentals including data gathering, validation, and processing. Additionally, how the results can be visualised and explained. They will also develop their skillset to study deployed Intelligent Systems and evaluate technical and ethical challenges and opportunities.

Learning Outcomes

By the end of this unit, students will be able to:

LO1 Discuss the theoretical foundation of Artificial Intelligence and its impact on users and organisations

LO2 Analyse the approaches, techniques and tools to deploy Intelligent Systems in an organisation

LO3 Modify an AI-based system to improve how exhibits intelligence in response to a real-world problem

LO4 Evaluate the technical and ethical challenges and opportunities of Intelligent Systems.

Essential Content

LO1 Discuss the theoretical foundation of Artificial Intelligence and its impact on users and organisations

AI Fundamentals:

Understanding what defines Artificial/Machine Intelligence; philosophical debates around the ambitions of simulating human intelligence; and the phenomenon of the “AI effect”.

AI and the phenomenon of combinatorial explosion.

The requirements of the underlying Data and its influence on AI outcomes.

How to handle large (big data) versus small datasets.

Understanding what “Learning from experience” means for Intelligent Agents and Intelligent Systems.

Appreciating the difference between AI and its subfields such as Machine Learning and related interdisciplinary research areas such as Robotics

How AI leverages other disciplines such as computer science, mathematics, psychology, software engineering, and linguistics.

Recognising traditional problems (goals) of AI such as reasoning, planning, learning, natural language processing, perception, prediction/forecasting, anomaly detection, computer vision, knowledge mining, and conversational AI.

Decision-making including basics of utility theory, sequential decision problems, elementary game theory, decision theory.

Understanding Intelligent agents: reactive, deliberative, goal-driven, utility-driven, and learning agents.

AI Types:

The difference between weak AI and strong AI.

Artificial Narrow Intelligence (ANI), also referred to as weak AI with applications focused on singular tasks e.g. Alexa, Siri, prediction tools, spam filters.

Artificial General Intelligence (AGI), also known as strong AI or deep AI e.g. Recognition, Recall, Hypothesis testing, Imagination, Analogy, Implication.

Artificial Super-intelligence (ASI), a hypothetical concept.

AI Applications:

The role of AI on the principles of a Universal Design.

Ambient Intelligence enabling electronic environments that are sensitive and responsive to the presence and preferences of people.

Finance e.g. to detect anomalies in charges outside of the norm, flagging these for human investigation.

Agriculture e.g. predicting the time it takes for a crop to be ripe and ready for picking, harvesting robot, predicting and extending storage and shelf life.

Business and eCommerce e.g. chatbots, visual searches, intelligent virtual assistants.

Engineering e.g. Computer Aided Design (CAD) and automation in factories.

Healthcare e.g. care of the elderly, heart beats analysis, computer-aided interpretation of medical images, drug discovery.

Cybersecurity e.g. profiling anomalous user behaviour, automating response against large-scale attacks.

Logistics and Supply Chain e.g. autonomous trucks and robotic picking systems.

Other examples include any application which exhibit intelligence via AI techniques such as strategy games, autopilot in autonomous cars, intelligent routing in computer networks, and military simulations.

LO2 Analyse the approaches, techniques and tools to deploy Intelligent Systems in an organisation

Approaches:

The major differences between top-down and bottom-up approaches to AI.

Explainable AI (XAI).

Statistical methods, computational intelligence, and traditional symbolic AI.

AI Tools, Libraries, Platforms, and Frameworks:

Options include but not limited to Tensorflow, Torch, Theano, Azure Machine Learning, Azure Cognitive Services, Azure Bot Service, MathWorks Matlab (plus Simulink), CNTK (Computational Network Toolkit), Deeplearning4j, Scikit-Learn, Swift AI IBM Watson, Keras, Pybrain, Google ML kit, Caffe, H20: open source AI platform.

Algorithms and techniques:

Understanding Machine Learning algorithms and processes including dataset preparation, feature engineering and selection, training and validating datasets, model training, selecting and interpreting model evaluation metrics and model deployment and management.

Linear Regression, Logistic Regression, Decision Tree, K-Nearest Neighbour, SVM (Support Vector Machine), Naïve Bayes, KNN (K- Nearest Neighbours), K-Means, Random Forest, Dimensionality Reduction Algorithms, Gradient Boosting & AdaBoost.

Tools and required relationships for testing, e.g. accurate and clear documentation, role of static testing and review in early defect detection, the need to follow specific industry standards (e.g. GDPR, health informatics, safety critical) and psychology mindset of tester-developer relationship.

LO3 Modify an AI-based system to improve how it exhibits intelligence in response to a real-world problem.

AI-based system

Common types of computer vision solution including image classification, object detection solutions, optical character recognition, facial detection, recognition and analysis.

Common types of natural language processing including key phrase extraction, entity recognition, language modeling, speech recognition and synthesis, and translation.

Common types of conversational AI e.g. webchat bots.

Modification:

Modify existing AI-based system using cloud based solutions e.g. Azure Machine Learning studio, Azure Cognitive Services and Azure Bot service.

Identifying the need to make modifications.

Modifying commands.

Impact of modification on cost and quality.

Improvement identification e.g. accuracy, efficiency, speed.

Application selection:

Criteria for AI-based application selection, e.g. any application software, system or agent which exhibits intelligence as part of its problem-solving approach e.g. open-source projects from Google and GitHub.

AI Analysis:

Overfitting, underfitting.

Data collection, data sources and assessment of data reliability to modify AI-based system.

LO4 Evaluate the technical and ethical challenges and opportunities of Intelligent Systems

Ethics in the Use of AI:

Identify guiding principles for responsible AI e.g. fairness, reliability, safety, privacy, security, inclusiveness, transparency and accountability.

Use of deep learning in recruiting new employees e.g. Deepfake.

AI bias and the ethical dilemma e.g. potential to widen socio-economic inequality, AI powered hiring processes (employment opportunities), access to skilling, health/life extension, algorithmic quantitative trading.

Autonomous weapons (mass casualties), AI arms race, Ethical implications of autonomous weapons.

Challenges:

Overfitting, AI lack of reasoning e.g. naïve physics, folk psychology.

The impact of data quality and quantity e.g. on the accuracy of an AI algorithm.

Job automation, risks of mass unemployment.

Intelligent Systems and Intelligent Agents have no emotions or out-of-the-box thinking.

Limited understanding of the AI decision making process e.g. deep learning.

Challenges related to the lack of compliance frameworks while considering legal and emerging legal factors e.g. GDPR, Data Protection and governance.

Risks; privacy and security e.g. Deepfake technology, emerging technology, aligning AI goals with objective(s).

Challenges related to readiness e.g. Lack of understanding of AI (and the value of data) among non-technical employees, lack of business alignment, robust testing, alignment of AI goals with defined objectives.

AI and Intelligent Systems are emerging technologies, not fully tested.

The environmental footprint of AI e.g. the carbon impact of AI.

Myth and fiction around AI e.g. mythical worry of “AI turning conscious” vs actual worry, “AI turning competence with objectives misaligned with ours”.

Opportunities:

Artificial cognitive abilities could make faster and more accurate decisions e.g. intelligence advice in health care.

Enabling affordability of services e.g. automation reduces operational costs.

Meeting demand e.g. the optimisation of routine processes increases productivity.

Inform strategic decision making e.g. profiling and risk assessment based on large datasets to predict high-risk events/actors.

Mitigate physical harm e.g. an AI-driven robot replaces a human in a dangerous location.

Availability an AI system can work 24×7.

Introducing new innovations e.g. AI as a competitor advantage (AI is an emerging technology with growing potential enabled by increasing processing power).

Collaborative work with human input e.g. AI and humans work together to reduce false positives.

Collaborative Robots (Cobots) and use in industry, healthcare, etc.

Learning Outcomes and Assessment Criteria

Recommended Resources

Textbooks

Géron, A. (2019). Hands-on machine learning with Scikit-Learn, Keras, and TensorFlow: Concepts, tools, and techniques to build intelligent systems. O’Reilly Media.

Lauterbach, A., & Bonime-Blanc, A. (2018). The artificial intelligence imperative: a practical roadmap for business. ABC-CLIO.

Liu, Y. (2019). Python Machine Learning By Example: Implement machine learning algorithms and techniques to build intelligent systems. 2nd edn. Packt Publishing.

Marcus, G., & Davis, E. (2019). Rebooting AI: Building artificial intelligence we can trust. New York, USA: Pantheon.

Russell, S., & Norvig, P. (2019). Artificial intelligence: a modern approach. 4th edn. Pearson.

Zaccone, G., & Karim, M. R. (2018). Deep Learning with TensorFlow: Explore neural networks and build intelligent systems with Python. Packt Publishing Ltd.

Links

This unit links to the following related units:

Unit 25: Machine Learning

Unit 46: Robotics

Unit 47: Emerging Technologies

Unit 48: Virtual and Augmented Reality Development.