Please enable JavaScript.
Coggle requires JavaScript to display documents.
Artificial Intelligence and Machine Learning - Coggle Diagram
Artificial Intelligence and Machine Learning
Privacy and Security Challenges
Scenario One
Employee Wellness Monitoring
Ethical Choice
Implement a wellness monitoring system that allows employees to opt-in voluntarily and anonymizes all data used to improve workplace health.
Unethical Choice
Require all employees to participate in wellness monitoring without consent, collecting personal health data that could later be used to evaluate their performance
Best Choice
Allowing voluntary participation respects autonomy, while anonymizing data minimizes privacy invasion. This aligns with Kantian ethics, where respecting autonomy is key to treating people with dignity.
Scenario Two
City Data Collection
Ethical Choice
Collect only anonymized, aggregated data in public spaces to improve city services while informing residents about the data collected and how it's used.
Unethical Choice
Collect detailed, identifiable data from residents through city sensors to create precise insights, but without clear public awareness or consent
Best Choice:
Ethical. Anonymizing data and providing transparency aligns with utilitarianism, aiming to maximize public benefit while respecting privacy.
Cybersecurity Advantages
Scenario One
Biometric Authentication in Public Services
Ethical Choice
Use biometric authentication only for high-security services, with explicit consent from users and alternatives for those uncomfortable with sharing biometric data.
Unethical Choice
Mandate biometric authentication for all services, without providing options for users who may have privacy concerns or disabilities that limit access.
Best Choice
Ethical. Offering alternatives ensures inclusivity and respects personal choice, supporting Rawl's principle of justice, which promotes fairness and access for all.
Scenario Two
Geo-Tracking for Emergency Response
Ethical Choice
Enable geo-tracking only during confirmed emergencies, using consent-based technology that alerts users when they are being tracked for safety.
Unethical Choice
Keep geo-tracking continuously active in all public safety systems, regardless of user consent, to monitor citizens.
Best Choice
Ethical. Using tracking only in emergencies and with consent respects autonomy and limits intrusion, in line with deontological ethics, which values actions based on principles like respect for autonomy.
Cybersecurity Risks
Scenario One
Discovering a Data Breach in a Competitor’s System
Ethical Choice
Quietly inform the competitor of the breach so they can address it and prevent harm to their users, while avoiding public disclosure that could cause panic.
Unethical Choice
Publicize the competitor’s data breach to gain a competitive advantage, regardless of the potential harm to their users
Best Choice
Ethical. Informing them quietly aligns with virtue ethics by showing integrity, and it prioritizes user's safety, following utilitarianism by aiming to minimize harm.
Scenario Two
Employee Phishing Simulation Test
Ethical Choice
Conduct phishing simulations with clear guidelines and training provided beforehand, so employees understand the purpose and feel respected, not tricked.
Unethical Choice
Run deceptive phishing tests without prior training or warning, then penalize employees who fail the test.
Best Choice
Ethical. Providing training upholds fairness, allowing employees to learn rather than be penalized, which aligns with fairness-focused ethical theories like Rawl's theory of justice.
Ethical Challenges
Scenario One
AI-Based Credit Scoring
Ethical Choice
Regularly audit the AI model to ensure it doesn’t use biased factors like race, gender, or location, and provide a transparent appeals process for customers.
Unethical Choice
Prioritize efficiency over fairness by using a credit scoring AI without transparency or accountability, even if it leads to discrimination.
Best Choice
Ethical. Auditing and offering appeals respects fairness and aligns with the consquentialist view, as it minimizes harm and promotes equity in financial services.
Scenario Two
Facial Recognition at Public Events
Ethical Choice
Limit facial recognition use to verifying ticketed attendees who consent, providing clear notice, and deleting the data immediately after the event.
Unethical Choice
Use facial recognition on all attendees without informing them, retaining data indefinitely for future marketing purposes.
Best Choice
Ethical. Informing and obtaining consent respects individual autonomy, and deleting data afterward limits privacy risks, which aligns with Kantian ethics in treating people as moral agents with rights to privacy.