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
Senario 1: A food delivery service is collecting additional data from the users without their consent
Option 1: The food delivery service can keep collecting dats however this would violate user right because they did not agree to additional tracking of data.
Option 2: The service couldInform users and require clear consent. Although they wouldn't get as much data they would more transparent in their actions and refrain form violating the users privacy.
Deontological Ethics - Respect for individuals
Senario 2: A phone operating system is selling messages to a third party
Option 1: The owners of the operating system could continue sell data without notice. This action would hurt trust with the customer and show a lack of respect for customer data.
Option 2: The operating system could deploy an Opt-in only for data sharing program. This action protect user data from outside entities if they wished to use it without consent.
Ethical Challenges
Scenario 2: An tool made for improving grammar is outputting incorrect sentences sentences.
Option 1: the AI decisions could be kept secret which would continue hidden injustices. Defendants and the public cannot challenge or understand AI errors. This will bred inequality and destroys trust in the legal system.
Option 2: By making the AI decision-making transparent,
institutions can prioritize relationships and responsibility. Transparency ensures vulnerable populations are protected and unjust outcomes are corrected publicly.
Justice Ethics - Fairness and equality
Ethics of Care - Relational ethics
Scenario 1: A companies AI hiring tool favors men over women.
Option 1: The company could Ignore biases however failure to address bias leads to systemic injustice and reinforces discrimination. It results in lost opportunities for qualified candidates and legal risks under anti-discrimination laws.
Option 2: Implementing regular algorithm audits would ensure fairness and impartiality. Regular audits catch biases early, create more equitable hiring processes, and enhance organizational credibility.
Cybersecurity Risks
Scenario 2: AI launches network attacks automatically.
Option 1: Having no human oversight removes moral accountability. Mistakes could trigger massive societal harm, which would be ethically indefensible.
Option 2: Implementing A human override would keep cyber tools accountable, preventing society from being harmed by systems that could develop their own agendas.
Utilitarianism - Greatest good for the greatest number
Social Contract Theory - Duty to society
Scenario 1: Hackers trick AI into false alerts Causing increased time in troubleshooting with in a network.
Option 1: This breaches could be ignored which could save money but leave sensitive data vulnerable to attacks from the outside.
Option 2: Strengthening AI resilience protects a larger population from harm and maintains safety within the AI developed.
Cybersecurity Advantages
Scenario 2: AI strengthens a banks cybersecurity by deploying serval layers of protection and flagging suspected threats found in the system.
Option 1: If the bank completely relies solely on AI, there is a potential that attacks occur on the AI and that would derail the whole system build for protection.
Option 2: By Combining AI defenses with human monitoring the system can optimize its protection services by minimizing false positives and incorrect identification of sensitive data.
Virtue Ethics - Wisdom and care
Consequentialism - Maximize positive outcomes
Scenario 1: An AI flags phishing emails for a company to prevent any cyber security attacks from happening.
Option 1: Applying fully automated response who save man power and time but could fall into unexpected issues that can be over looked by the AI.
Option 2: Having AI and a Human review insurers the greatest protections against threats because the AI can catches issues in the system that the human can not a vice versa.