CAT 2025 Slot 3 · VARC · RC passage 2: Can Ethics Be Formalised? AI and Moral Judgement

The full passage and all 4 questions (Q5–Q8) from CAT 2025 Slot 3, with answers and solutions.

The passage below is accompanied by four questions. Based on the passage, choose the best answer for each question.

Imagine a world in which artificial intelligence is entrusted with the highest moral responsibilities: sentencing criminals, allocating medical resources, and even mediating conflicts between nations. This might seem like the pinnacle of human progress: an entity unburdened by emotion, prejudice or inconsistency, making ethical decisions with impeccable precision. . . .

Yet beneath this vision of an idealised moral arbiter lies a fundamental question: can a machine understand morality as humans do, or is it confined to a simulacrum of ethical reasoning? AI might replicate human decisions without improving on them, carrying forward the same biases, blind spots and cultural distortions from human moral judgment. In trying to emulate us, it might only reproduce our limitations, not transcend them. But there is a deeper concern. Moral judgment draws on intuition, historical awareness and context – qualities that resist formalisation. Ethics may be so embedded in lived experience that any attempt to encode it into formal structures risks flattening its most essential features. If so, AI would not merely reflect human shortcomings; it would strip morality of the very depth that makes ethical reflection possible in the first place.

Still, many have tried to formalise ethics, by treating certain moral claims not as conclusions, but as starting points. A classic example comes from utilitarianism, which often takes as a foundational axiom the principle that one should act to maximise overall wellbeing. From this, more specific principles can be derived, for example, that it is right to benefit the greatest number, or that actions should be judged by their consequences for total happiness. As computational resources increase, AI becomes increasingly well-suited to the task of starting from fixed ethical assumptions and reasoning through their implications in complex situations.

But what, exactly, does it mean to formalise something like ethics? The question is easier to grasp by looking at fields in which formal systems have long played a central role. Physics, for instance, has relied on formalisation for centuries. There is no single physical theory that explains everything. Instead, we have many physical theories, each designed to describe specific aspects of the Universe: from the behaviour of quarks and electrons to the motion of galaxies. These theories often diverge. Aristotelian physics, for instance, explained falling objects in terms of natural motion toward Earth's centre; Newtonian mechanics replaced this with a universal force of gravity. These explanations are not just different; they are incompatible. Yet both share a common structure: they begin with basic postulates – assumptions about motion, force or mass – and derive increasingly complex consequences. . . .

Ethical theories have a similar structure. Like physical theories, they attempt to describe a domain – in this case, the moral landscape. They aim to answer questions about which actions are right or wrong, and why. These theories also diverge and, even when they recommend similar actions, such as giving to charity, they justify them in different ways. Ethical theories also often begin with a small set of foundational principles or claims, from which they reason about more complex moral problems.

Question 5 · RC Main Idea and Central Theme

Which one of the options below best summarises the passage?

  1. A

    The passage weighs the appeal of an impersonal AI judge against doubts about moral grasp. It warns that codification can erode case-sensitive judgement, allow axiom-led reasoning at scale, and use a physics analogy to model structured plurality.

  2. B

    The passage rejects formal methods in principle. It holds that moral judgement cannot be expressed in disciplined terms and concludes that AI should not serve in courts, medicine, or diplomacy under any conditions.

  3. C

    The passage weighs the appeal of an impersonal AI judge against doubts about moral grasp. It claims codified schemes retain case nuance at scale and uses a physics analogy to predict convergence on a unified framework.

  4. D

    The passage highlights administrative gains from automation. It treats reproducing human moral judgement as progress and argues that, as computational resources increase, AI can be responsible for decision-making across varied institutional settings.

Pick an option, then check.
Answer & solution

Answer: Option A

Map the passage before comparing options. Paragraph 1: the appeal of an AI moral arbiter, free of emotion and prejudice. Paragraph 2: the doubt – moral judgement draws on intuition, history and context, which resist formalisation, so encoding it risks flattening what matters. Paragraph 3: attempts at formalisation anyway, with utilitarianism's axiom and AI reasoning from fixed assumptions at scale. Paragraphs 4 and 5: the physics analogy – many incompatible theories sharing a common postulate-and-derivation structure, and ethics being similar.

Note what the physics analogy actually establishes. Physics has no single theory explaining everything; the theories diverge and are incompatible, yet share a structure. The analogy models plurality with shared structure, not convergence.

Option 1 carries all four movements and gets the analogy right – 'structured plurality'. It is the answer.

Option 3 is identical to option 1 in its opening and then fails twice. It says codified schemes 'retain case nuance at scale', the reverse of paragraph 2's warning, and it says the physics analogy predicts 'convergence on a unified framework', the reverse of 'There is no single physical theory that explains everything'.

Option 2 overstates the scepticism. The passage does not reject formal methods in principle – paragraph 3 begins 'Still, many have tried to formalise ethics' and treats that seriously – nor does it conclude that AI should never serve.

Option 4 drops the doubt altogether and treats reproducing human judgement as progress, whereas paragraph 2 says reproducing our limitations is precisely the worry.

Shortcut: when two options share an opening sentence, the discrimination lies entirely in what follows. Compare only the differing clauses against the text rather than rereading the passage.

Hence, the answer is option 1.

Question 6 · RC Vocabulary in Context

Choose the one option below that comes closest to being the opposite of "utilitarianism".
  1. A
    The policy was cast as deontological ethics, selecting the option that delivered the highest total benefit to citizens while presenting duty as a secondary consideration in public decision-making.
  2. B
    The authors advocated an absolutist stance, following exceptionless rules regardless of outcomes and evaluating choices by broadest societal benefit.
  3. C
    The committee adopted a non-egoist framework, ranking policies by their contribution to overall social welfare and treating self-interest as a derivative concern within institutional evaluation.
  4. D
    The council followed a prioritarian approach, assigning greater moral weight to improvements for the worst-off rather than to maximising total welfare across the affected population.
Pick an option, then check.
Answer & solution

Answer: Option D

Take the definition from the passage itself: utilitarianism 'often takes as a foundational axiom the principle that one should act to maximise overall wellbeing'. So the opposite must be an approach that does not aim at maximising total welfare.

Here is the trap in this question. Each option names a label – deontological, absolutist, non-egoist, prioritarian – and then describes what was actually done. Read the description, not the label.

Option 1 calls itself deontological but describes 'selecting the option that delivered the highest total benefit to citizens'. That is utilitarian behaviour wearing a different name.

Option 2 calls itself absolutist and speaks of exceptionless rules regardless of outcomes, but then says choices are evaluated 'by broadest societal benefit'. The second half cancels the first and lands back on maximising total welfare.

Option 3 ranks policies by 'contribution to overall social welfare', which is the utilitarian criterion itself.

Option 4 is the only description that departs from the criterion: greater moral weight to the worst-off 'rather than to maximising total welfare'. Distribution, not the total, decides. That is the opposite of utilitarianism.

Shortcut: in options built as label-plus-description, cover the label and judge by the description alone. Three descriptions here say 'highest total benefit', 'broadest societal benefit' and 'overall social welfare' – the same idea thrice. The odd description is the answer.

Hence, the answer is option 4.

Question 7 · RC Inference Questions

All of the following can reasonably be inferred from the passage EXCEPT:
  1. A
    with fixed moral starting points and expanding computational resources, the argument forecasts convergence on one ethical system and treats contextual judgement as unnecessary once formal reasoning scales across domains and cultures.
  2. B
    encoding ethics into fixed structures risks stripping away intuition, history, and context and, if that occurs, the depth that enables reflective judgement disappears. So, machines would mirror our limits rather than exceed them.
  3. C
    by analogy with physics, compact postulates can yield broad predictions across incompatible theories and ethics can likewise share structure while continuing to diverge rather than close on a single comprehensive framework.
  4. D
    the appeal of an AI judge rests on immunity to bribery, partiality, and fatigue; yet the text questions whether procedural cleanliness amounts to moral understanding without lived context and interpretive depth.
Pick an option, then check.
Answer & solution

Answer: Option A

Three options are supportable; one is not. Check each against the text.

Option 2 restates paragraph 2 almost line for line – moral judgement draws on intuition, historical awareness and context, encoding it risks flattening its essential features, and AI would reproduce our limitations rather than transcend them.

Option 3 restates paragraphs 4 and 5 – many physical theories, incompatible yet sharing a postulate-and-derivation structure, with ethics said to have a similar structure while its theories also diverge.

Option 4 restates paragraph 1's appeal – an entity 'unburdened by emotion, prejudice or inconsistency' – together with paragraph 2's doubt about whether that amounts to understanding morality.

Option 1 cannot be inferred, and fails on two separate counts. The passage insists there is no single physical theory that explains everything and that ethical theories likewise diverge, so no convergence on one ethical system is forecast. And the passage treats context as essential and resistant to formalisation, so it never treats contextual judgement as unnecessary.

Shortcut: the passage's whole architecture is plurality plus irreducible context. Any option asserting convergence or the dispensability of context is contradicting the structure of the argument, not just a sentence of it.

Hence, the answer is option 1.

Question 8 · Reading Comprehension

The passage compares ethics to physics, where different theories apply to different aspects of a domain and says AI can reason from fixed starting points in complex cases. Which one of the assumptions below must hold for that comparison to guide practice?
  1. A
    There is a principled way to decide which ethical framework applies to which class of cases, so the system can select the relevant starting points before deriving a recommendation.
  2. B
    A single master framework replaces all others after translation into one code, so domain boundaries disappear in application.
  3. C
    Once formalised, all ethical frameworks yield the same recommendation in every case, so selection among them is unnecessary.
  4. D
    Real cases never straddle different areas, so a case always fits exactly one framework without any overlap whatsoever.
Pick an option, then check.
Answer & solution

Answer: Option A

Spell out the comparison. In physics there are many theories, 'each designed to describe specific aspects of the Universe', and they are incompatible with one another. Ethics is said to have a similar structure – several divergent theories, each starting from its own foundational principles.

Now add the practical claim: AI reasons from fixed starting points through to implications in complex situations. If there are several incompatible sets of starting points, the system must first know which set to use for the case in hand. Without that, the comparison cannot guide practice at all.

Option 1 states that requirement – a principled way to decide which framework applies to which class of cases, so the relevant starting points can be selected before deriving a recommendation. It is the assumption.

Option 2 collapses the plurality into one master framework. That destroys the analogy, which rests on many theories coexisting.

Option 3 says all frameworks yield identical recommendations. The passage says the opposite – ethical theories 'even when they recommend similar actions, such as giving to charity, they justify them in different ways', and they diverge.

Option 4 is far too strong. 'Never straddle' and 'without any overlap whatsoever' demand a tidiness the passage never claims; and the assumption needed is a way to choose, not a guarantee that choosing is always trivial.

Shortcut: a necessary assumption is the weakest statement that closes the gap. When one option says 'there is a principled way to decide' and another says 'cases never overlap', the modest one is almost always the assumption and the absolute one the trap.

Hence, the answer is option 1.

Take this paper as a timed test
Real exam timing, instant score, rank and topic-wise analysis.

Practice timed →