Should Machines Make Moral Choices? The Ethics of Technology
1 of 4
RE · My World Quest — full scheme of work · pioneer
Should Machines Make Moral Choices? The Ethics of Technology
3 more
1/4
For grown-ups
Companion summary
Have you ever wondered who is responsible when a self-driving car makes a wrong decision?
Every app you use was built by someone who made choices about fairness, privacy, and who gets left out.
In 2018, an AI hiring tool was scrapped after it was found to favour men over equally qualified women.
When you finish, try asking: should a machine ever make decisions that affect a person's whole future?
✨
A Driverless Dilemma
🌊🐟🚣💧
A Driverless Dilemma
Imagine a self-driving car travelling at speed. Its brakes fail. The car's computer must choose: swerve left and hit one pedestrian, or go straight and hit five. The car makes this decision in milliseconds — with no human hand on the wheel. Here is the striking part: nobody programmed the car to choose who lives. The programmers just wrote rules about physics and safety. Yet the outcome is a life-or-death moral choice made by a machine. Who is responsible? The programmer? The company? The owner? The car itself? Hold those questions in your mind — they are at the heart of today's lesson.
💭 Think about this
Think of one piece of technology you used today. Could it ever cause harm to someone, even accidentally? How would you feel if it did?
✨
A Driverless Dilemma
🌊🐟🚣💧
A Driverless Dilemma
Imagine a self-driving car travelling at speed. Its brakes fail. The car's computer must choose: swerve left and hit one pedestrian, or go straight and hit five. The car makes this decision in milliseconds — with no human hand on the wheel. Here is the striking part: nobody programmed the car to choose who lives. The programmers just wrote rules about physics and safety. Yet the outcome is a life-or-death moral choice made by a machine. Who is responsible? The programmer? The company? The owner? The car itself? Hold those questions in your mind — they are at the heart of today's lesson.
💭 Think about this
Think of one piece of technology you used today. Could it ever cause harm to someone, even accidentally? How would you feel if it did?
💡
Technology, Ethics, and Moral Responsibility
🔢➕✖️🧮
Technology is not morally neutral. Every tool humans build reflects choices — about what matters, who benefits, and who might be harmed. is the branch of philosophy and religious thought that asks: what is right, wrong, good, or harmful? When we apply ethics to technology, we call this the ethics of technology.
Three major ethical frameworks help us think this through.
1. judges actions by their outcomes. If a technology produces more benefit than harm overall, a consequentialist says it is acceptable. Facial recognition software, for example, can catch criminals — a benefit. But it also misidentifies innocent people and enables surveillance of whole populations — serious harms. The consequentialist asks: do the benefits outweigh the costs, and for whom?
2. (linked to thinkers like Immanuel Kant, and echoed in many religious traditions) judges actions by duties and rights, not results. A deontologist argues that some actions are simply wrong, regardless of their outcomes. Collecting someone's personal data without consent violates their right to privacy — full stop. No beneficial result justifies it.
3. asks: what would a person of good character do? Many faith traditions — from Buddhism's principle of ahimsa (non-harm) to Islamic concepts of maslaha (public interest) to Christian agape (selfless love) — align closely with virtue ethics. A virtue ethicist designing an would ask: am I acting with honesty, fairness, and compassion?
Technology also raises specific ethical issues that cut across all frameworks:
• and fairness: Algorithms trained on biased historical data reproduce that bias. AI hiring tools have been shown to favour certain groups, disadvantaging others.
• Privacy: Smart devices constantly collect data about where you are, what you buy, and what you say. Who owns that information?
• The digital divide: Not everyone has equal access to technology. This inequality affects education, health, and economic opportunity.
• Autonomous weapons: Military drones and AI-guided weapons can strike targets without a human pulling the trigger. This raises profound questions about moral accountability in warfare.
• Environmental impact: Data centres powering cloud services consume enormous amounts of electricity, contributing to climate change.
is shared. Developers, companies, governments, and users all play a role. Religious and humanist traditions consistently argue that being human means taking responsibility for the tools we create — because tools shape human dignity, justice, and community.
Worked example
SCENARIO: A healthcare company builds an AI system to predict which patients are most likely to need urgent hospital care. The AI was trained using historical medical records. Researchers later discover the system consistently under-rates the urgency of patients from lower-income backgrounds, because those patients historically received less medical care — so the data wrongly suggests they are healthier.
Applying the frameworks:
• Consequentialist view: The AI causes real harm — vulnerable patients receive less care, outcomes worsen, health inequalities deepen. Even if the overall system helps most patients, the harm to a specific group is serious enough to demand the system be corrected or withdrawn.
• Deontological view: Using a biased system violates patients' rights to equal treatment and dignity. It is wrong regardless of how many people it helps overall. Developers had a duty to test for fairness before deployment.
• Virtue ethics / religious perspective: A compassionate and just developer — one guided by principles like maslaha or agape — would have prioritised fairness from the start. Discovering the bias and continuing to use the system without fixing it would be a failure of moral character.
Who is responsible? The programmers who chose the training data; the company that approved deployment without adequate fairness testing; and the regulators who allowed it. Responsibility is distributed — not zero for anyone.
🧠
Test Your Thinking
Try it yourself
Read each question carefully. Choose the best answer. For Question 5, think through your reasoning before selecting.
1. A social media company collects users' private messages to sell to advertisers without telling users. Which ethical framework most directly condemns this as wrong, regardless of how much advertising revenue it raises?
2. An AI recruitment tool consistently rejects applications from women for engineering jobs because it was trained on historical data from a male-dominated industry. This is an example of which specific ethical issue in technology?
3. A consequentialist evaluating facial recognition technology in airports would primarily ask:
4. Which of the following best describes the concept of 'moral responsibility' when an autonomous weapon causes civilian casualties?
5. A tech company discovers its cloud servers consume vast amounts of electricity from fossil fuels, contributing to climate change. A Muslim engineer argues the company has a duty to switch to renewable energy because of the concept of maslaha. Which statement best explains their reasoning?
📌
Bringing It Together
Technology is never morally neutral — every design choice reflects values and creates ethical consequences.
Consequentialism, deontological ethics, and virtue ethics offer different but complementary lenses for evaluating technology.
Moral responsibility for technology's harms is always shared — between developers, companies, governments, and users.
1. A government proposes using AI to automatically assess welfare benefit claims, with no human review of the decisions it makes. Which pair of ethical concerns does this raise most directly?
🌱 Reflect
Think about one piece of technology you personally use. Using one of the three ethical frameworks from today, explain one question about that technology you think society should be asking — but probably isn't.
