Please enable JavaScript.
Coggle requires JavaScript to display documents.
AI Decision-Making in Moral Dilemmas - Coggle Diagram
AI Decision-Making in Moral Dilemmas
self-driving cars
Moral Dilemma: Who should the car prioritise in an unavoidable accident?
Deontological Approach: Follow strict ethical rules (e.g., always protect passengers).
Reality: AI relies on statistical risk assessments, not built-in ethics.
Utilitarian Approach: Minimise total harm (save the most lives).
Future Considerations: Should laws enforce ethical AI in transport?
Moral Dilemma: Should self-driving cars break traffic laws to prevent harm? Example: An AI-driven car detects a pedestrian crossing illegally.
Utilitarian Approach: Break the law (swerve into an empty lane) to prevent injury.
Deontological Approach: Follow the law and stop, even if a rear-end collision occurs.
Reality: AI prioritises legal compliance over human-like moral intuition.
Future Considerations: Should AI be allowed to adapt its decisions based on context?
Are AI Systems Ethical?
AI relies on statistical models, not moral reasoning.
Human oversight is essential to apply ethics in decision-making.
Ongoing Debate: Should AI be programmed with ethical rules or remain purely data-driven?
medical AI
Moral Dilemma: Who should get medical treatment first in an emergency?
Reality: AI suggests treatment priorities based on probability models, but human doctors override AI when ethics demand it.
Utilitarian Approach: Prioritise patients with the highest survival probability.
Deontological Approach: Treat all patients equally, regardless of prognosis.
Future Considerations: Can AI balance efficiency with fairness?
Moral Dilemma: Should AI reject a doctor’s ethical concerns? Example: A patient refuses treatment, but AI predicts they will die without it.
Utilitarian Approach: Disregard patient consent if survival chances are high.
Deontological Approach: Respect patient decision, even if it leads to harm.
Reality: AI provides recommendations, but final decisions remain human-controlled.
Future Considerations: Should AI have more authority in life-and-death cases?