○​ Real World Example: The driver decides to input the parking duration.

4.​ Performing the Action​

What can I do now?

○​ UI Example: The user clicks "Search Flights."

○​ Real World Example: The driver inserts coins or uses a card to pay.

5.​ Perceiving the State of the System​

What just happened?

○​ UI Example: A list of available flights is displayed.

○​ Real World Example: The kiosk displays a ticket confirmation.

6.​ Interpreting the State​

What does this feedback mean?

UI Example: The user understands that flights are available at specific times.​
Real World Example: The driver checks if the ticket confirms payment.

7.​ Evaluating the Outcome​

Did I accomplish my goal?

○​ UI Example: The user confirms that the desired flight is available.

○​ Real World Example: The driver places the ticket on their dashboard.

The Gulfs

1.​ Gulf of Execution

○​ Definition: The gap between a user’s intention and the system’s
available actions.

○​ Example in UI: A user struggles to figure out how to apply a discount

code in a payment form.

○​ Example in Real World: A driver can’t find the “pay” button on a

parking meter.​
TWO MARKERS
1. First-ever chatbot (ELIZA)

ELIZA was created by Joseph Weizenbaum in 1966 as one of the first chatbots. It used
pattern-matching to simulate human conversation, often responding with questions to
encourage further interaction.

2. What is Artificial Intelligence?

AI is the simulation of human intelligence in machines, allowing them to perform tasks like
problem-solving, learning, and decision-making.

3. Primary function of the chatbot ALICE

ALICE (Artificial Linguistic Internet Computer Entity) is a chatbot that uses natural language
processing to engage in human-like conversations. For example, it can respond to "How are
you?" with "I'm just a computer, but I'm doing well!"

4. Sensation vs. Perception

●​ Sensation: The raw data received by senses (e.g., seeing light).​

●​ Perception: The brain's interpretation of sensory input (e.g., recognizing a face).​

5. Father of AI (John McCarthy)

John McCarthy coined the term "Artificial Intelligence" in 1956. He developed LISP, one of
the first AI programming languages, and worked on early AI concepts.

6. Weak AI vs. Strong AI

●​ Weak AI: Performs specific tasks (e.g., Siri, Google Assistant).​

●​ Strong AI: Hypothetical AI with human-like intelligence.​

7. Reducing Cognitive Load for Air Traffic Controllers

1.​ Automated Alerts – Prioritize important information.​

2.​ Simplified Interface – Reduce clutter for quick decision-making.​

8. What are heuristics in design?

Heuristics are usability guidelines that help create user-friendly interfaces (e.g., “Provide
clear feedback” and “Prevent user errors”).

9. Error message for a failed monetary transaction

"Transaction failed: Insufficient funds. Please check your balance and try again."

10. What is reinforcement learning?

Reinforcement learning is a trial-and-error learning method where an AI system receives

rewards or penalties to optimize its behavior (e.g., AI playing chess and improving moves).

11. Supervised vs. Unsupervised Machine Learning

●​ Supervised: AI learns from labeled data (e.g., email spam detection).​

●​ Unsupervised: AI finds patterns in unlabeled data (e.g., customer segmentation).​

12. Two sources of human errors in Human-Machine Interaction (HMI)

1.​ Slip – Error in execution (e.g., mistyping a password). ​

2.​ Mistake – Wrong decision (e.g., choosing the wrong menu option).​

13. Two of Nielsen’s heuristics with examples

1.​ Visibility of system status – Progress bars for file uploads.​

2.​ Error prevention – Disabling the "Submit" button for incomplete forms.​

14. Impact of lack of situation awareness on decision-making

If a pilot fails to notice a weather alert due to distractions, they may make a poor decision,
endangering the flight.

15. Halo vs. Horn Effect in Design

●​ Halo Effect: A beautiful website is assumed to be well-functioning.​

●​ Horn Effect: A basic-looking app is assumed to be outdated, even if it works well.​

16. What is closure in interface design?

Closure is the principle where users perceive incomplete visuals as complete, such as a
dashed circle appearing whole.

17. Importance of Hick’s Law in UI

Hick’s Law states that more choices lead to slower decision-making. Limiting menu
options improves usability.

18. Human factors in young children (age 10) & UI design consideration

Children have developing attention spans—designs should use bright colors, larger
fonts, and simple interactions.

19. Importance of Universal Design

Universal design ensures accessibility for all users, including those with disabilities,
making interfaces more inclusive.

20. Two AI applications in aviation

1.​ Autopilot – AI assists pilots in navigation.​

2.​ Predictive Maintenance – AI detects mechanical issues before failures.​

21. Two ways to design for colorblind users

1.​ Use patterns and text labels instead of colors alone.​

2.​ Allow users to customize interface colors.​

22. Equity vs. Equality in Universal Design

●​ Equity: Adjusting designs to meet individual needs (e.g., adjustable text size).​

●​ Equality: Providing the same design for everyone (e.g., one default text size).​

23. Two ways design thinking affects behavior

1.​ Encourages empathy – Designers create user-centered solutions.​

2.​ Promotes creativity – Users engage more with well-designed interfaces.​

24. What is a prototype?

A prototype is a basic working model used for testing and refining designs before final
development.

10 MARKERS
1. Explain the different types of machine learning with relevant
examples. If you were tasked with developing a machine learning
algorithm to play a chess game, describe the type of machine learning
you would choose and justify your decision.

Machine Learning (ML) is categorized into three primary types:

1. Supervised Learning

●​ The AI is trained on labeled datasets, meaning input data has known outputs.​

●​ Example: Email spam detection—AI learns which emails are spam based on labeled
examples.​

●​ Pros: High accuracy, useful for classification and prediction tasks.​

●​ Cons: Requires a large amount of labeled data.​

2. Unsupervised Learning

●​ The AI is given unlabeled data and must find patterns on its own.​

●​ Example: Customer segmentation—AI groups customers based on purchasing

behavior without prior labels.​

●​ Pros: Can discover hidden structures in data.​

●​ Cons: Harder to evaluate and interpret results.​

3. Reinforcement Learning

●​ AI learns through trial and error, receiving rewards for correct actions and penalties
for mistakes.​

●​ Example: Self-driving cars—AI learns safe driving behavior by maximizing positive

outcomes (e.g., avoiding collisions).​

●​ Pros: Useful for decision-making and continuous learning.​

 ●​ Cons: Requires extensive computational power and time.​

Machine Learning for Chess AI

For a chess-playing AI, reinforcement learning is the best choice. The AI can play games
against itself, adjusting its moves based on rewards (winning) or penalties (losing).

A great example is AlphaZero, developed by DeepMind. Unlike earlier chess engines that
relied on human knowledge, AlphaZero played millions of games against itself, discovering
advanced strategies beyond human capabilities.

Thus, reinforcement learning is preferred because it enables the AI to continually improve

and develop new strategies beyond pre-programmed rules.

2. AI is categorized based on capabilities and functionalities. Examine

only those categories based on functionalities with relevant examples.

AI is classified into four functional categories:

1. Reactive Machines

●​ These AI systems do not store past experiences and react solely to the current
input.​

●​ Example: IBM’s Deep Blue chess engine—it evaluates millions of board positions
but does not learn from past games.​

2. Limited Memory AI

●​ These systems learn from historical data to improve decision-making.​

●​ Example: Self-driving cars—they analyze past road conditions, recognize patterns,

and make real-time decisions.​

3. Theory of Mind AI (Hypothetical)

●​ This AI would understand human emotions, intentions, and social interactions.​

●​ Example: An AI therapist capable of detecting emotional distress and responding

empathetically.​

4. Self-Aware AI (Hypothetical)
 ●​ This AI would possess consciousness and independent thought.​

●​ Example: No existing example, but this would be an AI capable of self-improvement

without human intervention, similar to how humans think.​

These categories determine how AI interacts with the world, shaping its potential impact
on industries like healthcare, finance, and robotics.

3. Review the complex ethical dilemmas that are likely to be encountered

in the field of artificial intelligence.

AI presents several ethical challenges due to its ability to process vast amounts of data,
automate tasks, and influence decision-making.

1. AI Bias & Discrimination

●​ AI systems inherit biases from human-generated training data.​

●​ Example: AI-based hiring tools have been shown to favor male candidates over
female ones due to biased training data.​

●​ Solution: Ensure diverse datasets and implement fairness algorithms.​

2. Privacy Violations

●​ AI-powered surveillance systems collect personal data without user consent.​

●​ Example: Facebook’s facial recognition technology raised concerns about privacy

breaches.​

●​ Solution: Enforce stricter data protection laws like GDPR.​

3. Autonomous Weapons & Security Risks

●​ AI can be used to develop military weapons that operate without human control.​

●​ Example: Lethal autonomous drones that target individuals based on AI-based

identification.​

●​ Solution: Establish international AI regulations to prevent misuse.​

4. AI & Job Displacement

 ●​ AI automates repetitive tasks, leading to job losses.​

●​ Example: AI-powered robots replacing human workers in manufacturing industries.​

●​ Solution: Governments should focus on reskilling displaced workers.​

5. AI Decision-Making & Accountability

●​ Who is responsible when an AI system makes a wrong decision?​

●​ Example: If a self-driving car causes an accident, should the manufacturer, the

developer, or the owner be held accountable?​

●​ Solution: Develop clear legal frameworks assigning responsibility.​

Addressing these ethical dilemmas is crucial to ensuring AI is used responsibly for the
benefit of society.

4. Explore any five Gestalt principles with examples and discuss how
these concepts can improve the human-machine interface.

Gestalt principles help designers create user-friendly and visually appealing interfaces
by explaining how humans perceive patterns.

1. Figure-Ground Principle

●​ Users naturally distinguish between foreground (important content) and

background.​

●​ Example: Darkening the background of a pop-up modal helps users focus on the
pop-up.​

2. Proximity Principle

●​ Objects placed close together are perceived as related.​

●​ Example: Grouping similar settings in a menu makes navigation easier.​

3. Similarity Principle

●​ Elements that look alike are perceived as part of the same group.​
 ●​ Example: Using the same button color for all "Submit" actions across a website.​

4. Continuity Principle

●​ Users prefer to follow continuous paths instead of disconnected elements.​

●​ Example: A website’s breadcrumb navigation (e.g., Home → Products → Checkout)

helps users follow a logical flow.​

5. Closure Principle

●​ The brain fills in missing parts of an incomplete shape.​

●​ Example: The dashed-line "loading" icon still appears as a full circle to users.​

How These Principles Improve Human-Machine Interaction:

1.​ Enhances user experience by reducing confusion.​

2.​ Improves accessibility by making elements intuitive.​

3.​ Speeds up navigation and task completion.​

4.​ Ensures consistency across different platforms.​

5.​ Encourages engagement by making interfaces visually appealing.​

By applying Gestalt principles, designers can create interfaces that feel natural and easy to
use, improving both usability and accessibility.

5. Describe the concept of deep learning with suitable examples. Explain

the difference between slips and mistakes with relevant examples.

Deep Learning

Deep learning is a subset of machine learning that uses artificial neural networks to
mimic the way the human brain processes information.

How Deep Learning Works

●​ A neural network consists of multiple layers that process data hierarchically.​

 ●​ Each layer extracts features (e.g., in facial recognition, layers detect edges, eyes,
and finally the full face).​

Examples of Deep Learning

1.​ Image Recognition – AI recognizes objects in photos (e.g., Google Photos tagging
people).​

2.​ Natural Language Processing (NLP) – AI-powered chatbots understand and

generate human speech (e.g., ChatGPT, Google Assistant).​

3.​ Healthcare – AI detects diseases from X-rays and MRIs with high accuracy.​

Slips vs. Mistakes in Human Errors

Slips (Execution Errors)

●​ Occur when a person intends to do the right thing but makes an unintended
action.​

●​ Example: Typing "1234" instead of "1243" in a password field.​

Mistakes (Decision Errors)

●​ It Happens when the intention itself is wrong due to poor decision-making.​

●​ Example: A pilot misinterprets a low-fuel warning and takes the wrong emergency
action.​

Key Difference:

●​ Slips = Wrong execution (e.g., pressing the wrong button).​

●​ Mistakes = Wrong intention (e.g., misunderstanding instructions).​

Understanding deep learning and human errors is crucial for designing AI systems that
reduce human mistakes and improve decision-making.

6. Explain the Turing Test approach within the realm of artificial

intelligence. Draw a flowchart outlining an online monetary transaction,
incorporating an error message designed to alert users and protect them
from falling victim to online scams.

The Turing Test

●​ Proposed by Alan Turing in 1950, the Turing Test evaluates whether a machine can
exhibit human-like intelligence.​

●​ If a human judge cannot distinguish between a conversation with an AI and a

human, the AI passes the test.​

Turing Test Process

1.​ A human judge communicates via text with both an AI and a human.​

2.​ If the judge cannot reliably tell which is the AI, the AI is considered intelligent.​

3.​ This test measures natural language processing, reasoning, and deception
abilities.​

Example of AI Passing the Turing Test

●​ In 2014, a chatbot named Eugene Goostman tricked 33% of judges into believing it
was a human.​

Flowchart for Online Monetary Transaction with Scam Protection

A secure online transaction should:

●​ Verify user identity.​

●​ Detect suspicious activities.​

●​ Provide clear error messages to warn users of potential fraud.​

Here's the step-by-step process:

[Start]

[User enters payment details]

 ↓

[System verifies transaction] → [Is payment flagged as suspicious?]

-- Yes --> [Show scam warning message] → [Ask for user confirmation]

↓ No

[Proceed with transaction]

[Transaction Successful / Failed Message]

[End]

Example of Scam Warning Message:

"Warning: This payment appears unusual. Please verify the recipient before proceeding. If
you suspect fraud, contact customer support immediately."

Such systems help reduce fraudulent transactions and improve user trust in online
payment platforms.

7. Discuss potential attentional errors such as change blindness,

inattentional blindness, tunnel vision, and selective attention. Share
suggestions for alleviating these challenges while developing a web
page.

1. Change Blindness

●​ What it is: Users fail to notice changes in an interface when they are not in focus.​

●​ Example: A button changes position after an update, but users still search for it in
the old location.​

●​ Fix: Use highlight animations or notifications to indicate changes.​

2. Inattentional Blindness

●​ What it is: Users miss obvious elements when focused elsewhere.​

 ●​ Example: A user ignores a security pop-up because they are rushing to click
"Submit."​

●​ Fix: Use timed delays or visual emphasis (e.g., bold colors) for critical alerts.​

3. Tunnel Vision

●​ What it is: Users focus too much on one area, ignoring surrounding details.​

●​ Example: A user filling out a form overlooks an important instruction at the top.​

●​ Fix: Place essential information near input fields or highlight key instructions.​

4. Selective Attention

●​ What it is: Users filter out less relevant information, sometimes missing important
details.​

●​ Example: A user ignores a help tooltip because they assume they already know how
to use the feature.​

●​ Fix: Introduce contextual prompts that appear only when needed.​

By addressing attentional errors, designers can create more intuitive and error-resistant
web pages.

8. Examine any five cognitive biases that can influence user behavior
and deploy strategies to counter their impacts when crafting an
interface.

1. Halo Effect

●​ What it is: A visually appealing design can make users assume the product
functions well, even if usability is poor.​

●​ Fix: Balance aesthetics with functional usability testing.​

2. Confirmation Bias

●​ What it is: Users seek information that supports their preexisting beliefs while
ignoring contradictory data.​
 ●​ Fix: Provide neutral feedback and diverse perspectives in decision-making tools.​

3. Availability Heuristic

●​ What it is: Users rely on recent or easily recalled experiences rather than complete
data.​

●​ Fix: Show real-time analytics instead of relying on recent feedback only.​

4. Anchoring Effect

●​ What it is: Users depend too much on initial information when making decisions.​

●​ Fix: Offer comparative options instead of focusing on a single "anchor" price or

feature.​

5. False Consensus Effect

●​ What it is: Designers assume users think like them, leading to unintuitive designs.​

●​ Fix: Conduct user testing with diverse audiences to validate design choices.​

By recognizing and countering these biases, UX designers can create more effective and
inclusive interfaces.

9. Appraise the similarities and differences between Nielsen’s heuristics

and Shneiderman's golden rules.

Similarities

1.​ Both focus on usability – Ensuring user interfaces are intuitive and efficient.​

2.​ Both emphasize error prevention – Encouraging error messages, undo options, and
clear feedback.​

3.​ Both promote user control – Allowing users to correct mistakes and navigate freely.​

Differences

Nielsen’s Heuristics Shneiderman’s Golden Rules

 10 usability principles 8 design rules

Focuses on recognition, flexibility, Focuses on shortcuts, reducing memory load,

and consistency and responsiveness

Broader scope (applies to various UIs) More structured and step-based approach

Both frameworks guide designers in creating user-friendly systems but Shneiderman’s

rules focus more on workflow efficiency, while Nielsen’s heuristics prioritize cognitive
ease.

10. Assess Nielsen’s 10 interface design principles while keeping in

mind the desired outcomes of each principle.

Heuristic Desired Outcome

1. Visibility of System Status Users should always know what is happening

(e.g., progress bars).

2. Match Between System & Real Use familiar concepts and language (e.g., trash
World bin icon for "Delete").

3. User Control & Freedom Allow undo/redo actions to prevent frustration.

4. Consistency & Standards Ensure UI elements function predictably across

screens.

5. Error Prevention Prevent errors before they happen (e.g., disable

invalid options).
 6. Recognition Over Recall Reduce memory load by making options visible
(e.g., dropdown menus).

7. Flexibility & Efficiency of Use Provide shortcuts for advanced users.

8. Aesthetic & Minimalist Design Remove unnecessary elements for a clean UI.

9. Help Users Recognize & Recover Provide clear, helpful error messages.
from Errors

10. Help & Documentation Offer easy-to-access guidance for users.

By applying these principles, designers can improve usability, efficiency, and user
satisfaction.

11. Donald Norman’s interaction model concentrates on the user’s view

of the interface. Examine this statement by explaining the Gulf of
Execution, Gulf of Evaluation, and the Seven Stages of Action.

Donald Norman’s Interaction Model focuses on how users interact with a system and how
designers can bridge the gap between user intentions and system responses. It
consists of two key concepts:

1. Gulf of Execution

●​ Represents the gap between what a user wants to do and how the system allows
them to do it.​

●​ Example: If a user wants to print a document but cannot find the "Print" button, they
experience a high Gulf of Execution.​

●​ Solution: Use clear navigation, visible buttons, and intuitive workflows to

reduce this gap.​

2. Gulf of Evaluation

●​ Represents the gap between what the system does and how well the user
understands it.​
 ●​ Example: A user clicks "Send" on an email but gets no confirmation message,
leading to uncertainty.​

●​ Solution: Provide immediate feedback (e.g., "Email Sent Successfully") to reduce

this gap.​

3. Seven Stages of Action (Norman’s Model)

The model breaks interaction into seven steps:

Stage Description Example

1. Goal What the user wants to achieve. A user wants to book a flight.

2. Plan Deciding how to achieve the The user decides to use an airline
goal. website.

3. Specify Choosing which actions to take. The user selects a flight date and
destination.

4. Perform Executing the chosen actions. The user clicks "Search Flights".

5. Observing the system’s The system shows flight options.

Perceive response.

6. Interpret Understanding the response. The user reads flight details and prices.

7. Evaluating if the goal was The user confirms their booking is

Compare achieved. successful.

How This Helps in UI/UX Design

●​ Designers must minimize the Gulf of Execution and Evaluation to make

interactions seamless.​
 ●​ Intuitive designs with clear navigation, feedback, and user guidance improve
usability.​

Thus, Norman’s model helps in creating user-friendly interfaces that align with user
expectations.

12. You are responsible for designing a mobile application to submit

attendance. Determine relevant examples of how to reduce the cognitive
load on users of this application, considering the components of
working memory.

Understanding Cognitive Load & Working Memory

●​ Cognitive load refers to the mental effort required to complete a task.​

●​ The three components of working memory are:​

1.​ Phonological Loop – Handles auditory/verbal information.​

2.​ Visuospatial Sketchpad – Processes visual/spatial elements.​

3.​ Central Executive – Directs attention and manages tasks.​

Strategies to Reduce Cognitive Load in an Attendance App

1. Minimize Data Entry

●​ Use one-tap check-ins (e.g., QR code scanning instead of manual entry).​

●​ Example: Instead of typing "Present," allow students to tap a "Mark Attendance"

button.​

2. Reduce Visual Clutter

●​ Keep the interface simple and distraction-free.​

●​ Example: Show only relevant details (e.g., today’s class schedule) instead of
unnecessary options.​

3. Provide Clear Feedback

●​ Give real-time confirmation messages after submission.​

 ●​ Example: "Attendance Marked Successfully" instead of making users guess.​

4. Use Recognition Over Recall

●​ Avoid making users remember steps or details.​

●​ Example: Autofill student ID based on login credentials.​

5. Design for Quick Navigation

●​ Use large buttons and intuitive icons to reduce time spent searching for functions.​

●​ Example: A simple home screen with “Mark Attendance” as the primary focus.​

By reducing cognitive load, the app can enhance usability and efficiency, ensuring
students can mark attendance quickly and accurately.

13. Devise an AI-enabled wellness app using design thinking principles

for implementation in a corporate wellness program.

Using design thinking (Empathize, Define, Ideate, Prototype, Test), let’s design
"WellnessAI", a corporate wellness app.

1. Empathize (Understanding the Users)

●​ Employees experience stress, burnout, and lack of motivation.​

●​ Conduct surveys & interviews to understand their health needs.​

2. Define (Identifying the Problem Statement)

"Employees need a personalized, AI-powered wellness solution to improve mental and

physical health while balancing work-life responsibilities."

3. Ideate (Brainstorming Solutions)

●​ Feature 1: AI-generated personalized wellness plans (e.g., stress management

techniques).​

●​ Feature 2: AI chatbot for mental health support.​

●​ Feature 3: Smart reminders for hydration, exercise, and meditation.​

4. Prototype (Building a Basic Model)

●​ Develop an MVP (Minimum Viable Product) with a simple UI.​

●​ Test features like workout tracking and AI-recommended relaxation exercises.​

5. Test (User Feedback & Improvements)

●​ Employees use the app for 1 month, providing feedback.​

●​ Improve usability, notifications, and AI recommendations based on user

behavior.​

This design-thinking approach ensures the AI wellness app is user-friendly, impactful,

and engaging.

15. Apply the concept of design thinking to create a personal growth app
for a first-year undergraduate student and specify the human factors you
will consider for this student while developing the app.

Step 1: Empathize

●​ First-year students face stress, time management iss

●​ ues, and difficulty adjusting to college life.​

Step 2: Define

"Students need an AI-powered personal growth app to help with study planning, mental
health, and productivity."

Step 3: Ideate

●​ AI-based study planner with personalized schedules.​

●​ Mental health chatbot for stress relief.​

●​ Habit-tracking system to monitor daily progress.​

Step 4: Prototype

●​ Build a simple UI with reminder notifications, self-assessment quizzes, and

AI-generated study tips.​
Step 5: Test & Improve

●​ Gather student feedback and refine features for ease of use, motivation, and
effectiveness.​

A well-designed app can empower students to manage academics, mental health, and
personal development effectively.

16. Explain the concept of Universal Design and evaluate its importance
in interface development.

What is Universal Design?

●​ Universal Design (UD) is an approach that ensures products, environments, and

interfaces are accessible to all people, regardless of age, ability, or disability.​

●​ It aims to create inclusive experiences without requiring special adaptations.​

7 Principles of Universal Design

1.​ Equitable Use – The design should be usable by people with diverse abilities.​ ​

○​ Example: Voice-controlled assistants (e.g., Siri, Alexa) help visually impaired

users.​

2.​ Flexibility in Use – Users should have multiple ways to interact with an interface.​

○​ Example: Websites offering both keyboard shortcuts and mouse

navigation.​

3.​ Simple and Intuitive Use – The interface should be easy to understand.​

○​ Example: Icons with labels for better navigation.​

4.​ Perceptible Information – Important information should be presented in multiple

ways.​

○​ Example: Closed captions on videos for hearing-impaired users.​

5.​ Tolerance for Error – The design should minimize the impact of mistakes.​

○​ Example: An "Undo" button for accidental deletions.​

6.​ Low Physical Effort – The interface should be comfortable to use.​

 ○​ Example: Auto-fill forms reduce typing effort.​

7.​ Size & Space for Approach & Use – The design should accommodate different
users.​

○​ Example: Large touch targets for people with motor impairments.​

Importance of Universal Design in Interface Development

1.​ Enhances Accessibility – People with disabilities can use the system easily.​

2.​ Improves Usability – A well-designed interface benefits all users, not just
disabled ones.​

3.​ Expands User Base – More inclusive design = more potential users.​

4.​ Increases Legal Compliance – Many countries have laws requiring accessible
digital interfaces.​

5.​ Boosts Business Reputation – Companies that prioritize accessibility gain positive
public perception.​

Conclusion:​
Universal Design is not just about disability—it ensures that all users can interact
efficiently with technology. Prioritizing inclusivity improves usability, user satisfaction,
and engagement.

17. Compare and contrast AI-based decision-making with human

decision-making.

Artificial Intelligence (AI) is increasingly being used to assist or replace human

decision-making in various fields, from healthcare to finance.

Key Differences Between AI & Human Decision-Making

Aspect AI Decision-Making Human Decision-Making

Speed Processes data instantly. Slower due to thinking time.

Bias Can inherit bias from training Influenced by emotions and
data. personal experiences.

Adaptability Learns from patterns but lacks Can think outside the box and
creativity. adapt.

Emotional Lacks emotions and makes Decisions often involve emotions

Influence purely logical choices. and intuition.

Error Rate Low when trained properly. Prone to human errors like fatigue
and oversight.

Advantages of AI Decision-Making

1.​ Processes Large Data Sets – AI analyzes massive amounts of data faster than
humans.​

2.​ Eliminates Emotional Bias – AI makes purely data-driven decisions.​

3.​ Improves Efficiency – AI automates repetitive tasks, freeing up human resources.​

4.​ 24/7 Availability – Unlike humans, AI doesn’t need rest or breaks.​

Advantages of Human Decision-Making

1.​ Understands Context – Humans can interpret social and ethical nuances.​

2.​ Can Think Creatively – AI lacks innovation and imagination.​

3.​ Morality & Ethics – Humans consider the ethical implications of decisions better
than AI.​

AI + Human Hybrid Decision-Making

●​ Best Solution: AI-assisted human decision-making combines AI’s speed and

accuracy with human ethical reasoning.​

●​ Example: AI diagnoses diseases, but a human doctor makes the final treatment
decision.​
Conclusion:​
AI is great for data-driven tasks, but human intuition, ethics, and adaptability remain
irreplaceable. The best results come from AI and humans working together.