Cloud Computing & AI

Cloud Computing Fundamentals, Virtualization, Hypervisors, and Cloud Service Models

What Is Cloud Computing?

Cloud computing is the delivery of computing resources over the Internet instead of relying solely on local computers or on-premises hardware.

Rather than purchasing, maintaining, and upgrading expensive physical equipment, organizations can access computing resources as needed through cloud service providers.

Cloud resources may include:

Users typically access cloud resources through an Internet connection.

Why Organizations Use Cloud Computing

Cloud computing offers organizations greater flexibility and reduces the need for costly infrastructure.

Common reasons organizations move to the cloud include:

Cloud computing allows businesses to focus more on their operations rather than managing IT infrastructure.

Characteristics of Cloud Computing

Cloud services generally provide several important characteristics.

These include:

These characteristics make cloud computing flexible and cost-effective.

Advantages of Cloud Computing

Cloud computing provides many benefits.

Advantages include:

Organizations can increase or decrease resources as their needs change.

Disadvantages of Cloud Computing

Despite its advantages, cloud computing also presents challenges.

Potential disadvantages include:

Organizations should carefully evaluate these risks before migrating critical systems.

Virtualization

Virtualization is the technology that allows multiple virtual computers to operate on a single physical computer.

Instead of dedicating one physical machine to one operating system, virtualization enables several operating systems to share the same hardware.

Each virtual system behaves as though it is an independent computer.

Virtualization is one of the foundational technologies that makes cloud computing possible.

Benefits of Virtualization

Virtualization offers numerous advantages:

Organizations commonly virtualize servers to maximize efficiency.

Virtual Machine (VM)

A Virtual Machine (VM) is a software-based computer that runs inside another computer.

A virtual machine contains:

To users, a virtual machine appears and functions like a physical computer.

Host Machine

The host machine is the physical computer that provides hardware resources for virtual machines.

The host supplies:

The host may support multiple virtual machines simultaneously.

Guest Operating System

A guest operating system is an operating system installed inside a virtual machine.

Each guest operating system operates independently of the others.

Examples include:

Each guest receives only the resources allocated to it by the virtualization software.

Hypervisor

A hypervisor is software that creates, manages, and runs virtual machines.

It controls how physical hardware resources are shared among guest operating systems.

The hypervisor allocates:

Without a hypervisor, virtualization would not be possible.

Hypervisor Responsibilities

A hypervisor performs many important tasks.

These include:

Proper resource allocation ensures that multiple virtual machines can run efficiently.

Type 1 Hypervisor

A Type 1 hypervisor runs directly on physical hardware.

Characteristics include:

Examples include:

Type 1 hypervisors are commonly used in data centers and cloud environments.

Type 2 Hypervisor

A Type 2 hypervisor runs as an application on top of an existing operating system.

Characteristics include:

Examples include:

Type 2 hypervisors are common in educational and development environments.

Cloud Deployment Models

Cloud services can be deployed in several different ways depending on an organization’s requirements.

The primary deployment models include:

Public Cloud

A public cloud is owned and operated by a cloud provider.

Resources are shared among multiple customers.

Advantages include:

Public clouds are widely used by businesses of all sizes.

Private Cloud

A private cloud is dedicated to a single organization.

It may be located:

Advantages include:

Private clouds are often used by organizations with strict security or compliance requirements.

Hybrid Cloud

A hybrid cloud combines public and private cloud environments.

Organizations may store sensitive workloads in the private cloud while using the public cloud for less sensitive applications.

Benefits include:

Hybrid cloud deployments are becoming increasingly common.

Cloud Service Models

Cloud computing services are delivered using several service models.

The three primary models are:

Infrastructure as a Service (IaaS)

Infrastructure as a Service (IaaS) provides virtualized computing infrastructure.

Customers receive:

Customers manage:

The cloud provider manages the underlying physical infrastructure.

Platform as a Service (PaaS)

Platform as a Service (PaaS) provides both infrastructure and a software development platform.

Developers can focus on writing applications without managing servers or operating systems.

The provider manages:

Customers manage:

Software as a Service (SaaS)

Software as a Service (SaaS) provides complete software applications over the Internet.

Users simply access the application through a web browser or installed client.

Examples include:

The cloud provider manages nearly everything, while users simply use the application.

Comparing Cloud Service Models
Service ModelCustomer ManagesProvider Manages
IaaSOS, applications, dataHardware, networking, virtualization
PaaSApplications and dataInfrastructure, OS, runtime
SaaSApplication usage and dataEntire platform and application

Understanding these differences is important for selecting the appropriate cloud solution.

Best Practices

Key Terms

Exam Tips

Cloud Storage, Cloud Security, Artificial Intelligence Fundamentals, and Machine Learning

Cloud Storage

Cloud storage is a service that allows users and organizations to store data on remote servers that are accessible over the Internet.

Instead of saving files only on a local hard drive, users can store documents, images, videos, databases, and backups in the cloud.

Cloud storage providers maintain the physical storage infrastructure while customers access their files through applications or web browsers.

Benefits of Cloud Storage

Cloud storage offers many advantages.

These include:

Cloud storage is commonly used by individuals, schools, and businesses.

Common Uses of Cloud Storage

Organizations use cloud storage for:

Many cloud applications automatically save user data in cloud storage.

Cloud Scalability

One of the major advantages of cloud computing is scalability.

Scalability refers to the ability to increase or decrease computing resources based on current demand.

Organizations can:

This flexibility helps reduce unnecessary costs.

Elasticity

Elasticity is closely related to scalability.

Elastic cloud services automatically adjust available resources based on workload.

Examples include:

Elasticity allows organizations to pay only for the resources they actually use.

High Availability

High availability ensures that cloud services remain operational with minimal downtime.

Cloud providers achieve high availability through:

High availability is especially important for mission-critical applications.

Fault Tolerance

Fault tolerance allows systems to continue operating even if individual hardware components fail.

Examples include:

Fault-tolerant systems reduce the likelihood of service interruptions.

Disaster Recovery

Cloud providers often include disaster recovery capabilities.

Disaster recovery focuses on restoring systems and data after unexpected events.

Examples of disasters include:

Cloud-based disaster recovery reduces recovery time compared to traditional methods.

Cloud Security

Although cloud providers manage much of the infrastructure, customers still have security responsibilities.

Cloud security includes protecting:

Effective cloud security requires cooperation between the provider and the customer.

Shared Responsibility Model

Cloud security follows a Shared Responsibility Model.

Generally:

Cloud Provider Responsibilities

The provider secures:

Customer Responsibilities

Customers secure:

The exact division of responsibility depends on whether the service is IaaS, PaaS, or SaaS.

Identity and Access Management (IAM)

Identity and Access Management (IAM) controls who can access cloud resources.

IAM helps organizations:

Proper IAM configuration is essential for cloud security.

Multi-Factor Authentication (MFA)

Cloud accounts should always use Multi-Factor Authentication (MFA) whenever possible.

MFA combines two or more authentication factors, such as:

MFA significantly reduces the risk of unauthorized account access.

Encryption in the Cloud

Cloud providers use encryption to protect data.

Two important types include:

Data at Rest

Protects stored information.

Examples include:

Data in Transit

Protects information while it travels across networks.

Examples include:

Encryption helps maintain confidentiality and privacy.

Risks of Cloud Computing

Although cloud computing offers many advantages, organizations should understand potential risks.

These include:

Risk management should be part of every cloud deployment.

Artificial Intelligence (AI)

Artificial Intelligence (AI) refers to computer systems designed to perform tasks that normally require human intelligence.

Rather than simply following fixed instructions, AI systems can analyze data, identify patterns, make decisions, and improve performance over time.

Characteristics of Artificial Intelligence

AI systems may be capable of:

Different AI systems specialize in different tasks.

Traditional Programming vs. Artificial Intelligence

Traditional software follows predefined instructions written by programmers.

The process is generally:

Input → Program → Output

Artificial Intelligence differs because it learns patterns from data instead of relying solely on fixed rules.

AI systems improve as they process more training data.

Machine Learning (ML)

Machine Learning (ML) is a branch of artificial intelligence in which computers learn from data without being explicitly programmed for every possible situation.

Instead of manually writing every rule, developers provide:

The model discovers patterns on its own.

Machine learning powers many modern AI applications.

How Machine Learning Works

Machine learning generally follows these steps:

  1. Collect data.
  2. Prepare and clean the data.
  3. Train the model.
  4. Test the model.
  5. Evaluate accuracy.
  6. Deploy the model.
  7. Continue improving through additional training.

The quality of the data strongly influences the quality of the model.

Training Data

Training data is the information used to teach an AI model.

Examples include:

The more accurate and representative the training data, the better the model typically performs.

AI Models

An AI model is the trained system that performs predictions or generates responses.

After training, the model can:

Models improve through additional training and refinement.

Importance of High-Quality Data

AI systems depend heavily on the quality of their training data.

Poor-quality data can produce:

A common saying in data science is:

Garbage in, garbage out (GIGO).

Poor input data usually results in poor AI performance. AI models can also experience AI drift, where performance changes over time due to poor data or unintended model behavior, making regular evaluation and retraining important.

Responsible AI

Organizations should develop and use AI responsibly.

Responsible AI includes:

Developers should regularly test AI systems for accuracy and unintended bias.

Best Practices

Key Terms

Exam Tips

Generative AI, AI Applications, AI Ethics, and Responsible AI

Generative Artificial Intelligence (Generative AI)

Generative Artificial Intelligence (Generative AI) is a type of AI that creates new content rather than simply analyzing existing information.

Unlike traditional AI systems that classify data or make predictions, generative AI produces original outputs based on patterns learned during training.

Generative AI can create:

Large Language Models (LLMs) are one of the most common technologies used in generative AI.

How Generative AI Works

Generative AI is trained using extremely large collections of data.

The process generally includes:

  1. Collecting massive datasets.
  2. Cleaning and organizing the data.
  3. Training the AI model.
  4. Learning relationships and patterns.
  5. Generating new content from user prompts.

The AI does not memorize every answer. Instead, it predicts the most likely output based on what it learned during training.

Prompts

A prompt is the instruction or question given to an AI system.

The quality of the prompt often determines the quality of the AI’s response.

Examples include:

Well-written prompts usually produce more accurate and useful responses.

Prompt Engineering

Prompt engineering is the practice of designing prompts that guide an AI model toward producing the desired output.

Effective prompts often include:

Prompt engineering is becoming an important skill when working with AI systems.

Large Language Models (LLMs)

A Large Language Model (LLM) is an AI model trained on enormous amounts of text.

LLMs are designed to understand and generate natural language.

Common capabilities include:

LLMs predict the most probable sequence of words based on the user’s prompt.

AI Chatbots

An AI chatbot is a conversational program that communicates with users through text or voice.

Modern AI chatbots use natural language processing (NLP) to understand user requests and generate responses.

Common uses include:

AI chatbots can operate continuously without human intervention.

AI Assistants

An AI assistant is an intelligent software application designed to help users complete tasks.

Unlike simple chatbots, AI assistants often integrate with calendars, email, documents, and productivity tools.

Examples of tasks include:

AI assistants improve productivity by automating routine work.

AI-Generated Text

Generative AI can create written content such as:

Although AI-generated text can be highly useful, it should always be reviewed for accuracy before publication.

AI-Generated Images

AI systems can generate original images from text descriptions.

Applications include:

Image generation models learn artistic styles, objects, colors, and relationships from training data.

AI-Generated Audio

Generative AI can produce:

Applications include:

AI-Generated Video

AI can also create videos by generating or combining:

Businesses increasingly use AI-generated videos for marketing and employee training.

AI-Generated Code

Generative AI can assist software developers by producing:

Developers should always review AI-generated code because it may contain errors, security vulnerabilities, or inefficient logic.

Predictive AI

Predictive AI analyzes historical data to forecast future outcomes.

Rather than creating new content, predictive AI identifies patterns and estimates what is likely to happen next.

Examples include:

Predictive AI focuses on making informed predictions using existing data.

Generative AI vs. Predictive AI
Generative AIPredictive AI
Creates new contentPredicts future outcomes
Produces text, images, audio, codeProduces forecasts and classifications
Uses promptsUses historical data
Supports creativity and automationSupports decision-making and planning

Understanding the difference is important for selecting the appropriate AI solution.

Natural Language Processing (NLP)

Natural Language Processing (NLP) enables computers to understand, interpret, and generate human language.

NLP supports:

Modern AI assistants rely heavily on NLP technologies.

Computer Vision

Computer Vision enables AI systems to analyze and interpret images and videos.

Applications include:

Computer vision combines AI with image processing techniques.

Robotics and AI

Artificial intelligence enhances robotics by enabling machines to:

Examples include:

Common Applications of AI

Artificial intelligence is used in many industries.

Examples include:

Healthcare

Finance

Retail

Education

Transportation

Manufacturing

AI Limitations

Although AI is powerful, it has important limitations.

AI may:

Users should verify important AI-generated information before relying on it.

AI Hallucinations

An AI hallucination occurs when an AI system generates information that appears convincing but is inaccurate, fabricated, or unsupported.

Hallucinations may include:

Users should always fact-check AI-generated content, especially in academic, medical, legal, and business contexts.

Bias in AI

AI systems learn from existing data.

If the training data contains bias, the AI model may also produce biased results.

Bias may affect:

Developers should regularly evaluate AI systems for fairness.

Responsible AI

Responsible AI focuses on developing and using AI systems safely, ethically, and transparently.

Key principles include:

Organizations should establish governance policies for AI use.

AI Ethics

Ethical AI development seeks to ensure that AI benefits society while minimizing harm.

Important ethical considerations include:

Ethics should guide AI development throughout the system’s lifecycle.

Human Oversight

Humans remain responsible for decisions made with AI assistance.

Human review is especially important for:

AI should support—not replace—critical human decision-making.

Best Practices

Key Terms

Exam Tips

AI Security, Deepfakes and Governance

AI Security

As Artificial Intelligence becomes more widely used, protecting AI systems is becoming increasingly important.

AI security involves protecting:

Organizations must secure AI systems throughout their entire lifecycle.

Why AI Security Is Important

AI systems often process sensitive information.

Examples include:

A compromised AI system could expose confidential data or generate incorrect results.

Threats to AI Systems

AI systems face many of the same threats as traditional computer systems, along with several AI-specific risks.

Common threats include:

Understanding these threats helps organizations implement stronger security controls.

Data Poisoning

Data poisoning occurs when attackers intentionally insert false or misleading information into an AI model’s training data.

As a result, the model may:

Organizations should carefully validate training datasets before using them.

Prompt Injection

Prompt injection is an attack in which malicious instructions are designed to manipulate an AI model into ignoring its intended behavior or revealing information it should not disclose.

Possible consequences include:

Developers should validate user input and implement safeguards to reduce the risk of prompt injection.

Model Theft

Training advanced AI models requires significant time, expertise, and computing resources.

Model theft occurs when attackers copy or steal proprietary AI models without authorization.

Stolen models may be:

Protecting AI models is an important part of an organization’s intellectual property strategy.

Adversarial Attacks

An adversarial attack attempts to fool an AI model by making small, carefully crafted changes to input data.

For example:

Although the changes may be nearly invisible to humans, they can significantly affect AI predictions.

Deepfakes

A deepfake is AI-generated or AI-modified media that realistically imitates a person’s appearance, voice, or actions.

Deepfakes can include:

While they can be used for entertainment or education, they can also be used to spread misinformation, commit fraud, or impersonate individuals.

Risks of Deepfakes

Potential risks include:

Users should verify the authenticity of digital media before trusting or sharing it.

Synthetic Data

Synthetic data is artificially generated information created to resemble real-world data.

Organizations use synthetic data to:

When properly generated, synthetic data can help improve privacy while supporting AI development.

AI Governance

AI governance refers to the policies, processes, and controls used to manage AI systems responsibly.

Effective governance helps organizations ensure that AI systems are:

Governance establishes accountability throughout the AI lifecycle.

Components of AI Governance

A comprehensive AI governance program often includes:

Organizations should evaluate AI systems continuously rather than only during development.

Privacy and AI

AI systems often process personal information.

Organizations should protect privacy by:

Privacy should be considered throughout the design and deployment of AI systems.

Intellectual Property

AI-generated content may involve intellectual property considerations.

Organizations should:

Responsible use of AI includes respecting the rights of content creators.

Human Review

Although AI can automate many tasks, human review remains essential.

People should evaluate AI-generated outputs for:

Human oversight helps reduce the impact of AI errors.

Cloud and AI Working Together

Cloud computing and AI are closely connected.

Cloud platforms provide the computing power, storage, and networking resources needed to:

Many organizations use cloud-based AI services because they eliminate the need to build expensive AI infrastructure.

Benefits of Cloud-Based AI

Using AI in the cloud provides several advantages:

Cloud providers make advanced AI capabilities accessible to organizations of all sizes.

Challenges of Cloud-Based AI

Organizations should also consider potential challenges, including:

Planning and governance help address these challenges.

Cloud and AI Best Practices

To securely and effectively use cloud computing and AI:

Key Terms

Exam Tips

Cloud Computing

Artificial Intelligence

Responsible AI

AI Security