unit 1 revisiting ai project cycle ethical frameworks for ai
Subjective
ASSERTION AND REASON QUESTIONS
1. Assertion (A): The Data Exploration stage involves representing data visually.
Reason (R): Data collected is often large and difficult to analyze without visualization.
b. Both A and R are true, but R is not the correct explanation of A.
c. A is true, but R is false.
d. A is false, but R is true.
2. Assertion (A): A price comparison website belongs to the Computer Vision domain.
Reason (R): Statistical Data processes large numerical datasets.
b. Both A and R are true, but R is not the correct explanation of A.
c. A is true, but R is false.
d. A is false, but R is true.
3. Assertion (A): Ethical frameworks are important for AI decision-making.
Reason (R): Ethical frameworks prevent unintended harm.
b. Both A and R are true, but R is not the correct explanation of A.
c. A is true, but R is false.
d. A is false, but R is true.
4. Assertion (A): Justice and Beneficence are the same principle.
Reason (R): Beneficence promotes positive outcomes.
b. Both A and R are true, but R is not the correct explanation of A.
c. A is true, but R is false.
d. A is false, but R is true.
5. Assertion (A): Deployment is the final stage of the AI Project Cycle.
Reason (R): Evaluation happens before Deployment.
b. Both A and R are true, but R is not the correct explanation of A.
c. A is true, but R is false.
d. A is false, but R is true.
6. Assertion (A): Rights-based framework is sector-based.
Reason (R): Bioethics is sector-based.
b. Both A and R are true, but R is not the correct explanation of A.
c. A is true, but R is false.
d. A is false, but R is true.
7. Assertion (A): AI Project Cycle is cyclical.
Reason (R): It can be repeated to improve the solution.
b. Both A and R are true, but R is not the correct explanation of A.
c. A is true, but R is false.
d. A is false, but R is true.
8. Assertion (A): The case study violated Justice principle.
Reason (R): The model was trained on biased data.
b. Both A and R are true, but R is not the correct explanation of A.
c. A is true, but R is false.
d. A is false, but R is true.
9. Assertion (A): NLP analyzes visual information.
Reason (R): NLP understands human language.
b. Both A and R are true, but R is not the correct explanation of A.
c. A is true, but R is false.
d. A is false, but R is true.
10. Assertion (A): Data Acquisition is the most crucial stage.
Reason (R): Without data, no AI project can be built.
b. Both A and R are true, but R is not the correct explanation of A.
c. A is true, but R is false.
d. A is false, but R is true.
11. Assertion (A): Rights-based framework prioritizes human dignity.
Reason (R): It prevents violation of fundamental rights.
b. Both A and R are true, but R is not the correct explanation of A.
c. A is true, but R is false.
d. A is false, but R is true.
12. Assertion (A): "My Goodness" activity reveals personal biases.
Reason (R): It presents a moral dilemma influenced by personal values.
b. Both A and R are true, but R is not the correct explanation of A.
c. A is true, but R is false.
d. A is false, but R is true.
13. Assertion (A): Email spam filter is Computer Vision.
Reason (R): Computer Vision analyzes images and videos.
b. Both A and R are true, but R is not the correct explanation of A.
c. A is true, but R is false.
d. A is false, but R is true.
14. Assertion (A): Modelling stage tests model on new data.
Reason (R): Most efficient model is chosen in Modelling.
b. Both A and R are true, but R is not the correct explanation of A.
c. A is true, but R is false.
d. A is false, but R is true.
15. Assertion (A): Justice ensures fair distribution of benefits and burdens.
Reason (R): It ensures impartiality regardless of background.
b. Both A and R are true, but R is not the correct explanation of A.
c. A is true, but R is false.
d. A is false, but R is true.
16. Assertion (A): Virtue-based framework focuses on character and intentions.
Reason (R): It aligns actions with honesty and integrity.
b. Both A and R are true, but R is not the correct explanation of A.
c. A is true, but R is false.
d. A is false, but R is true.
17. Assertion (A): Ethical frameworks are only for complex AI projects.
Reason (R): All AI projects can cause harm.
b. Both A and R are true, but R is not the correct explanation of A.
c. A is true, but R is false.
d. A is false, but R is true.
18. Assertion (A): AI Project Cycle is cyclical.
Reason (R): Deployment is the final stage.
b. Both A and R are true, but R is not the correct explanation of A.
c. A is true, but R is false.
d. A is false, but R is true.
19. Assertion (A): Basic ethics knowledge is prerequisite.
Reason (R): It helps understand ethical frameworks.
b. Both A and R are true, but R is not the correct explanation of A.
c. A is true, but R is false.
d. A is false, but R is true.
20. Assertion (A): Data Exploration follows Data Acquisition.
Reason (R): Data must be collected before analysis.
b. Both A and R are true, but R is not the correct explanation of A.
c. A is true, but R is false.
d. A is false, but R is true.
Revisiting AI Project Cycle – Question & Answer Format
1. Outline the main steps in the AI Project Cycle briefly.
Answer:
- Problem Scoping: Define the problem clearly and identify the parameters affecting it.
- Data Acquisition: Collect relevant and authentic data from reliable sources.
- Data Exploration: Represent data visually using graphs and charts to identify patterns.
- Modelling: Select and build a suitable AI model to solve the problem.
- Evaluation: Test the model using new data and measure its efficiency.
- Deployment: Integrate the AI solution into real-world environments.
2. What roles does Computer Vision play in agricultural monitoring systems?
Answer:
- Crop Health Monitoring: Detect crop diseases and nutrient deficiencies.
- Automated Harvesting: Guide robotic systems to pick fruits and vegetables efficiently.
3. Mention the factors that knowingly or unknowingly influence our decision-making.
Answer:
- Non-maleficence: Avoid causing harm.
- Maleficence: Intentionally causing harm.
- Beneficence: Promoting well-being and positive outcomes.
4. What is the necessity for ethical frameworks in AI development?
Answer:
- Respect for Autonomy: Users should understand how AI systems function.
- Do Not Harm: Avoid causing harm to humans or non-humans.
- Maximum Benefit: Ensure AI provides maximum benefit.
- Justice: Ensure fair distribution of benefits and burdens.
5. Mention the key characteristics of sector-based frameworks.
Answer:
Sector-based frameworks are designed for specific industries. They address issues such as patient privacy, data security, and ethical AI use in sectors like healthcare, finance, education, transportation, agriculture, governance, and law enforcement.
6. What do you mean by Bioethics?
Answer:
Bioethics is an ethical framework used in healthcare and life sciences. It deals with ethical issues related to medicine, health, and biological sciences and ensures AI applications follow ethical standards.
7. What is Natural Language Processing? Explain any two real-life applications.
Answer:
Natural Language Processing (NLP) is a branch of AI that enables machines to understand, interpret, and process human language.
- Email Spam Filters: Detect spam messages based on certain words or patterns.
- Machine Translation: Systems like Google Translate convert text from one language to another.
8. How do value-based frameworks contribute to ethical decision-making?
Answer:
Value-based frameworks focus on fundamental ethical principles and moral values. They assess the moral worth of actions and guide responsible decision-making.
9. Define Artificial Intelligence and explain how machines become intelligent with an example.
Answer:
Artificial Intelligence refers to machines that can mimic human intelligence by making decisions, predicting outcomes, learning from data, and improving over time.
Example: Netflix recommends shows based on user preferences and updates suggestions when preferences change.
10. Explain the 4W Problem Canvas and Problem Statement Template.
Answer:
4W Problem Canvas:
- Who: People affected by the problem.
- What: Nature of the problem.
- Where: Situation or location where the problem occurs.
- Why: Benefits of solving the problem.
Problem Statement Template:
After filling the 4Ws, all points are summarized into a single structured statement describing the problem clearly.
11. How do you identify whether a machine/application is AI-based?
Answer:
A machine is AI-based if it is trained using data and can make decisions or predictions independently.
Example: A chatbot that understands language is AI, while a simple automation script is not.
12. After Problem Scoping, what are the remaining stages of the AI Project Cycle?
Answer:
- Data Acquisition
- Data Exploration
- Modelling
- Evaluation
- Deployment
13. Define Artificial Intelligence and list the three domains of AI.
Answer:
- Data Science: Uses numeric and alphanumeric data.
- Computer Vision: Uses images and videos.
- Natural Language Processing: Uses text and speech.
14. CBT Case Study – Answer the following.
Problem Statement Template:
People face stress due to competition and personal issues. AI-based CBT systems can help identify emotional patterns and provide support.
Two Data Sources:
- Surveys
- Interviews
How to Explore Data?
- Clean textual data
- Normalize text
- Extract important keywords
15. Create a 4W Project Canvas for Cybersecurity Scenario.
- Who: Organizations and network users.
- What: Cybersecurity risks and unusual data activity.
- Where: Network logs and digital systems.
- Why: Prevent cyber attacks and protect sensitive data.
AI Project Cycle Application:
- Data Acquisition – Collect log data.
- Data Exploration – Identify unusual patterns.
- Modelling – Use anomaly detection models.
- Evaluation – Measure accuracy.
- Deployment – Implement security system.
AI Project Cycle & Ethical Frameworks – 2 Marks Questions and Answers (Crash Course)
1. What are the two main types of ethical frameworks for AI?
Ans. The two main types are sector-based frameworks and value-based frameworks. Sector-based frameworks are tailored to specific industries like healthcare, while value-based frameworks focus on fundamental ethical principles that guide decision-making.
2. Define a framework in the context of problem-solving.
Ans. A framework is a set of steps that provides a structured, step-by-step guide for solving problems in an organized manner. They offer a common language for communication and collaboration, which helps in consistent problem-solving methodologies.
3. What is the primary objective of the Natural Language Processing (NLP) domain?
Ans. The primary objective of NLP is to enable machines to read, decipher, and understand human languages. It deals with the interaction between computers and humans using natural language, attempting to extract valuable information from spoken and written words using algorithms.
4. How does the Data Exploration stage differ from the Data Acquisition stage in the AI Project Cycle?
Ans. The Data Acquisition stage involves collecting data from various sources to form the foundation of the project. In contrast, the Data Exploration stage focuses on interpreting the collected data by giving it a visual representation, such as graphs or flow charts, to identify patterns.
5. Give two examples of applications that fall under the Computer Vision domain.
Ans. Two examples are agricultural monitoring and surveillance systems. In agriculture, drones with cameras capture images to assess crop health. In surveillance, computer vision is used to monitor public spaces and detect suspicious activities.
6. What are the key prerequisites for a student learning about ethical frameworks for AI?
Ans. The prerequisites are having a basic knowledge of AI and a basic understanding of ethics and ethics in AI. This foundational knowledge helps students grasp the concepts and apply the frameworks effectively.
7. Briefly explain the principle of "Respect for Autonomy" in bioethics.
Ans. Respect for Autonomy means ensuring users are fully aware of decision-making processes. For an AI algorithm, this means users should know how it works and have access to the data on which the models were trained.
8. What is the purpose of a case study at the end of the module on ethical frameworks?
Ans. The purpose is to demonstrate how to apply an ethical framework to an AI solution in a real-world scenario. The case study helps to show the impact of using such a framework to avoid unintended consequences.
9. What is the difference between "Non-maleficence" and "Maleficence"?
Ans. Non-maleficence is the ethical principle of avoiding causing harm or negative consequences. In contrast, Maleficence refers to the concept of intentionally causing harm or wrongdoing.
10. Mention two real-life applications of Natural Language Processing (NLP).
Ans. Two applications are email filters and machine translation systems. Email filters use NLP to identify and uncover spam messages, while machine translation systems like Google Translate use it to automatically translate text from one language to another.
11. Explain the role of the Evaluation stage in the AI Project Cycle.
Ans. The Evaluation stage involves testing the developed AI model on newly fetched data. The results of this testing help to evaluate the model's performance and provide insights for improving it.
12. What does the term "Beneficence" mean in the context of bioethics?
Ans. "Beneficence" is the ethical principle of promoting and maximizing the well-being and welfare of individuals and society. It emphasizes taking actions that produce positive outcomes and ensuring the greatest benefit is achieved for all stakeholders.
13. Why is the Deployment stage considered crucial for AI solutions?
Ans. The Deployment stage is crucial because it ensures the successful integration and operation of AI solutions in real-world environments. This is the stage where the solution begins to deliver value and impact to users and stakeholders.
14. What are the three categories of value-based frameworks?
Ans. The three categories are Rights-based, Utility-based, and Virtue-based frameworks. They reflect different moral philosophies that guide ethical reasoning.
15. What is the central focus of a Virtue-based ethical framework?
Ans. A Virtue-based framework focuses on the character and intentions of the individuals involved in decision-making. It asks whether actions align with virtuous principles like honesty, compassion, and integrity.
16. How do ethical frameworks help to avoid unintended consequences in AI solutions?
Ans. By using ethical frameworks during the AI solution-building process, developers can ensure that the AI makes morally acceptable choices. This systematic approach can help avoid unintended negative outcomes even before they occur.
17. What is the primary function of a Statistical Data system in AI?
Ans. The primary function is to collect numerous data, maintain data sets, and derive meaning or sense out of them. The information extracted is then used to make a decision.
18. Why is it important for the data used to train an AI model to be unbiased?
Ans. Unbiased data is crucial because if the training data is biased, the resulting AI algorithm may inadvertently exacerbate existing biases or inaccuracies. This can lead to flawed and harmful outcomes, as seen in the case study provided.
19. What is the ethical principle of "Justice" as applied to AI solutions?
Ans. The principle of Justice requires that all benefits and burdens of a particular AI solution or choice must be distributed in a justified manner across people, irrespective of their background, race, or social status.
20. Give one reason why an AI Project Cycle is useful.
Ans. The AI Project Cycle is useful because it provides an appropriate, structured framework that can lead a project towards its goal in an organized manner. It makes the execution of tasks clearer in the mind of the developer.
AI Project Cycle & Ethical Frameworks
3 Marks Questions with Answers
- Problem Scoping: Defining the problem and goal of the project.
- Data Acquisition: Collecting data from reliable sources.
- Data Exploration: Analyzing and visualizing data to find patterns.
- Modelling: Building and testing different models to find the most efficient one.
- Evaluation: Testing the chosen model on new data to assess its performance.
- Deployment: Integrating the final AI solution into a real-world environment.
- Statistical Data: Systems that collect and derive meaning from numerical or statistical datasets. Example: price comparison website.
- Computer Vision: Systems that analyze and interpret visual information from images and videos. Example: surveillance system.
- Natural Language Processing: Systems that process and understand human languages (spoken and written). Example: email spam filter.
- Sector-based framework: Tailored to specific sectors or industries, such as Bioethics for healthcare.
- Value-based framework: Focuses on fundamental ethical principles irrespective of the sector. Examples include Rights-based, Utility-based, and Virtue-based frameworks.
Example: A hiring algorithm biased against women could reinforce discrimination if ethical considerations are ignored.
In the case study, the AI algorithm violated this principle by recommending a less-ill group of patients for intensive care while harming more-ill patients from the western region by misallocating healthcare resources.
- Rights-based: Protects human rights and dignity.
- Utility-based: Maximizes overall good for the greatest number.
- Virtue-based: Focuses on character and intentions of decision-makers.
Factors influencing decisions include identity and location of the charity recipient, bias toward relatives, and availability of information.
- Data Acquisition: Collecting data from reliable sources.
- Data Exploration: Visual representation of collected data to interpret patterns.
AI Project Cycle & Ethical Frameworks – Long Answer Questions
1. Elaborate on the significance of the AI Project Cycle. Why is it important to follow these steps?
Answer:
The AI Project Cycle is a cyclical six-stage framework providing structured development.
- Ensures clarity in Problem Scoping.
- Builds strong foundation through Data Acquisition and Data Exploration.
- Enhances efficiency through Modelling and Evaluation.
- Ensures real-world impact through Deployment.
Without this structure, projects risk failure, inefficiency, and negative outcomes.
2. Discuss the importance of ethical frameworks in AI.
Answer:
Ethical frameworks are essential as AI is a powerful decision-making tool.
- Provide guidance in moral dilemmas.
- Help prevent bias.
- Align actions with virtuous principles.
- Allow proactive handling of ethical concerns.
They ensure moral acceptability of AI solutions.
3. Analyze the ethical considerations of the case study using "Respect for Autonomy" and "Justice".
Answer:
Respect for Autonomy: Violated as patients were unaware of how AI influenced their care. Transparency and accessibility of model data are required.
Justice: Violated due to biased data causing unfair resource allocation. A just solution must actively counter social biases and ensure equitable distribution.
4. Describe the three value-based frameworks and apply them to social media content moderation.
Answer:
- Rights-based: Protects freedom of expression while preventing rights violations like harassment.
- Utility-based: Removes harmful content to maximize overall good.
- Virtue-based: Focuses on building AI aligned with honesty, compassion, and fairness.
Each framework evaluates moderation decisions from a different moral perspective.
5. Explain the six stages of the AI Project Cycle with a real-world example.
Answer:
Example: AI-powered recipe recommendation app.
- Problem Scoping: Reduce food waste via recipe suggestions.
- Data Acquisition: Collect ingredient and recipe datasets.
- Data Exploration: Analyze ingredient patterns.
- Modelling: Train recommendation algorithm.
- Evaluation: Test with users and refine.
- Deployment: Release app for public use.