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 1: A hospital uses AI to analyze data, is it appropriate to share this data with a third party AI company?
Option 1: Use third party company without consent from patients
Option 2: Seek consent from patients for use of AI analysis with their sensitive health data
I think the option to seek consent from patients is important and aligns with Kantian ethics as it respects patients as ends in themselves
Scenario 2: Social Media uses AI to analyze private messages for advertisements
Option 1: Analyze private messages for advertisements
Option 2: Decline to use private messages for advertisements and instead focus on more public data
I think this option is the correct choice because it aligns with virtue ethics and respects individuals rights to privacy
Cybersecurity Risks
Scenario 1: Scammers use AI to mimic the voice of a CEO to confirm a transaction
Option 1: Rely on Multi factor authentication for transactions
This option would be the most ethical as it follows virtue ethics and supports individual responsibility
Option 2: Rely on voice recognition only
Scenario 2: The military develops a ground breaking AI model and does not know if they should release it
Option 1: Release model to public for open source innovation
Option 2: Keep model proprietary with strict security
I think this option would be most ethical, as releasing it could allow for military data to be reverse engineered and accessed. This follows deontological ethics and its duty to protect sensitive data
Ethical Challenges
Scenario 1: A company uses AI to screen new hires
Option 1: Use past hiring data with no further tweaks
Option 2: Tweak some past hiring data and use some audits to ensure performance
I think it is very important to alter some of the data so it does not reinforce any biases that past hiring techniques may have had. This follows utilitarianism at it can create the greatest good for the greatest number
Scenario 2: A school monitors school devices with AI to detect problematic student behavior
Option 1: Inform all parties clearly and implement strict boundaries
While it may be on school devices, I believe it is important to be transparent on privacy matters. This follows deontological ethics and respects individuals right to privacy
Option 2: Use the AI model without student consent as it is on school devices
Cybersecurity Advantages
Scenario 1: Accounting firm utilizes AI to detect potential fraudulent transactions
Option 1: Allow the AI to monitor every account for suspicious transactions
Option 2: Monitor only high risk accounts rather than all accounts
I think choosing to monitor only high risk accounts limits data access and provides the greatest benefit with the smallest risk in alignment with Utilitarianism.
Scenario 2: A hospital uses AI to detect suspicious logins, and blocks access if so.
Option 1: Keep auto blocks of suspicious logins implemented
Option 2: Implement an exception protocol for time sensitive logins
This option is the most ethical choice as it follows Kantian ethics and respects the professional judgement of doctors that may need rapid information