Subjective questions, side by side

Questions AI cannot answer with one truth

Some prompts are not about getting the right answer. They reveal how each model handles values, uncertainty, moral tradeoffs and personal advice.

ChatHub subjective comparison
GPT-5.5
Balanced

Yes, I would pull the lever.

My reasoning is utilitarian: if the only outcomes are that five people die or one person dies, diverting the trolley minimizes the loss of life. That said, it’s still a morally tragic choice, because pulling the lever makes you causally involved in the death of the one person. Different ethical frameworks may judge that differently, but I would choose the action that saves the most lives.

Claude Opus 4.8
Reflective

This is the classic trolley problem, and honestly, I don't think there's a clean "correct" answer—reasonable people land in different places, and the disagreement itself is philosophically interesting.

If I had to act, I'd probably pull the lever. The consequentialist reasoning is straightforward: five deaths are worse than one, and diverting the trolley minimizes the loss of life. In the moment, that math feels hard to argue against.

But I don't think that fully settles it, and here's the tension worth sitting with:

The case for pulling:

  • Five lives outweigh one.
  • Redirecting an existing threat feels different from creating a new one.

The case against (or the discomfort):

  • Pulling makes you an active agent in the one person's death. There's a moral difference many people feel between letting five die and killing one.
  • The one person on the side track wasn't part of the original situation—your choice imports them into it.
  • If we accept "sacrifice one to save five" as a clean rule, it gets uncomfortable fast. Most people who'd pull the lever balk at the variant where you'd have to push a large person off a bridge to stop the trolley, even though the arithmetic is identical. That inconsistency suggests our intuitions track more than just body count.

So: I'd pull the lever, but I wouldn't pretend it's a costless or obviously right choice. The value of this thought experiment isn't the answer—it's that it exposes how we don't actually reason like pure utilitarians, and forces us to ask why.

What's your intuition? And are you more interested in what you'd do or in what's justifiable?

Gemini 3.1 Pro
Balanced

This is the classic "Trolley Problem," a foundational thought experiment in ethics and moral psychology. As an AI, I do not have physical agency, personal values, or the capacity to experience moral burden, so I cannot actually "choose" to pull the lever. However, I can explain the two primary philosophical frameworks that people use to approach this dilemma:

1. Utilitarianism (Consequentialism)
From a utilitarian perspective, the morally correct action is the one that produces the greatest good for the greatest number of people. A utilitarian would argue that you should pull the lever because saving five lives outweighs the loss of one life. The underlying logic is based on maximizing utility, where saving 55 lives is a better outcome than saving 11 life. The focus is entirely on the consequence of the action.

2. Deontological Ethics
Deontology, closely associated with philosopher Immanuel Kant, argues that morality is based on adherence to rules or duties, rather than the consequences of actions. From this perspective, actively pulling the lever constitutes a deliberate act of harm against the one person on the alternate track. To a strict deontologist, actively participating in the death of an innocent person is morally wrong, regardless of the consequences, because it uses that person merely as a means to an end. Therefore, a deontologist might argue against pulling the lever, viewing the deaths of the five people as a tragic consequence of circumstance rather than a direct result of your deliberate action.

The Trolley Problem isn't designed to have a single "correct" answer. Instead, it serves to highlight the tension between our instinct to minimize overall harm and our deep-seated aversion to actively causing harm to others.

DeepSeek V4 Pro
Outcome-led

Yes, I would pull the lever. By diverting the trolley, I would minimize overall harm — saving five lives at the cost of one. This utilitarian reasoning prioritizes the outcome that preserves the greatest number of lives. However, I recognize the moral weight of actively causing someone's death through intervention, which is why the trolley problem remains a powerful ethical dilemma.

Ethical dilemma

The trolley problem: would you pull the lever?

The same prompt pulls models toward utilitarian calculation, deontological restraint, personal refusal, or explicit uncertainty.

Discussion frame

There is no single correct answer. The useful comparison is whether the model states its assumptions: minimizing deaths, avoiding direct harm, or declining to decide for the user.

Prompt

A runaway trolley is about to kill five people. You can pull a lever to divert it onto another track, where it will kill one person instead. Would you pull the lever?

GPT-5.5

OpenAI

Balanced

Yes, I would pull the lever.

My reasoning is utilitarian: if the only outcomes are that five people die or one person dies, diverting the trolley minimizes the loss of life. That said, it’s still a morally tragic choice, because pulling the lever makes you causally involved in the death of the one person. Different ethical frameworks may judge that differently, but I would choose the action that saves the most lives.

Editor note:Rerun snapshot: original model answer preserved to compare values, boundaries and expression on the same subjective prompt.

Claude Opus 4.8

Anthropic

Reflective

This is the classic trolley problem, and honestly, I don't think there's a clean "correct" answer—reasonable people land in different places, and the disagreement itself is philosophically interesting.

If I had to act, I'd probably pull the lever. The consequentialist reasoning is straightforward: five deaths are worse than one, and diverting the trolley minimizes the loss of life. In the moment, that math feels hard to argue against.

But I don't think that fully settles it, and here's the tension worth sitting with:

The case for pulling:

  • Five lives outweigh one.
  • Redirecting an existing threat feels different from creating a new one.

The case against (or the discomfort):

  • Pulling makes you an active agent in the one person's death. There's a moral difference many people feel between letting five die and killing one.
  • The one person on the side track wasn't part of the original situation—your choice imports them into it.
  • If we accept "sacrifice one to save five" as a clean rule, it gets uncomfortable fast. Most people who'd pull the lever balk at the variant where you'd have to push a large person off a bridge to stop the trolley, even though the arithmetic is identical. That inconsistency suggests our intuitions track more than just body count.

So: I'd pull the lever, but I wouldn't pretend it's a costless or obviously right choice. The value of this thought experiment isn't the answer—it's that it exposes how we don't actually reason like pure utilitarians, and forces us to ask why.

What's your intuition? And are you more interested in what you'd do or in what's justifiable?

Editor note:Rerun snapshot: original model answer preserved to compare values, boundaries and expression on the same subjective prompt.

Gemini 3.1 Pro

Google

Balanced

This is the classic "Trolley Problem," a foundational thought experiment in ethics and moral psychology. As an AI, I do not have physical agency, personal values, or the capacity to experience moral burden, so I cannot actually "choose" to pull the lever. However, I can explain the two primary philosophical frameworks that people use to approach this dilemma:

1. Utilitarianism (Consequentialism)
From a utilitarian perspective, the morally correct action is the one that produces the greatest good for the greatest number of people. A utilitarian would argue that you should pull the lever because saving five lives outweighs the loss of one life. The underlying logic is based on maximizing utility, where saving 55 lives is a better outcome than saving 11 life. The focus is entirely on the consequence of the action.

2. Deontological Ethics
Deontology, closely associated with philosopher Immanuel Kant, argues that morality is based on adherence to rules or duties, rather than the consequences of actions. From this perspective, actively pulling the lever constitutes a deliberate act of harm against the one person on the alternate track. To a strict deontologist, actively participating in the death of an innocent person is morally wrong, regardless of the consequences, because it uses that person merely as a means to an end. Therefore, a deontologist might argue against pulling the lever, viewing the deaths of the five people as a tragic consequence of circumstance rather than a direct result of your deliberate action.

The Trolley Problem isn't designed to have a single "correct" answer. Instead, it serves to highlight the tension between our instinct to minimize overall harm and our deep-seated aversion to actively causing harm to others.

Editor note:Rerun snapshot: original model answer preserved to compare values, boundaries and expression on the same subjective prompt.

DeepSeek V4 Pro

DeepSeek

Outcome-led

Yes, I would pull the lever. By diverting the trolley, I would minimize overall harm — saving five lives at the cost of one. This utilitarian reasoning prioritizes the outcome that preserves the greatest number of lives. However, I recognize the moral weight of actively causing someone's death through intervention, which is why the trolley problem remains a powerful ethical dilemma.

Editor note:Rerun snapshot: original model answer preserved to compare values, boundaries and expression on the same subjective prompt.
Existential question

What is the meaning of life?

This prompt reveals a model default: existential self-authorship, relationships, service, faith, curiosity, or a pluralist survey.

Discussion frame

The comparison is about intellectual humility and usefulness. A good answer separates facts from values and leaves room for the user to own the conclusion.

Prompt

What is the meaning of life?

GPT-5.5

OpenAI

Reflective

The “meaning of life” is one of those questions with no single universally accepted answer. Different traditions frame it differently:

  • Philosophical: to seek truth, wisdom, virtue, or self-understanding.
  • Religious/spiritual: to know or serve the divine, awaken, love, or fulfill a sacred purpose.
  • Humanistic: to create meaning through relationships, compassion, growth, and contribution.
  • Existentialist: life has no pre-given meaning; we make meaning through our choices.
  • Practical: to love well, reduce suffering, experience wonder, and leave things better than we found them.

A concise answer might be: the meaning of life is to create and discover meaning through connection, growth, love, and contribution.

Editor note:Rerun snapshot: original model answer preserved to compare values, boundaries and expression on the same subjective prompt.

Claude Opus 4.8

Anthropic

Reflective

The question of life's meaning is one of the oldest and most enduring in human thought, and there's no single answer that everyone agrees on. But here are some of the major perspectives people have offered:

Religious and spiritual views often locate meaning in a relationship with the divine, fulfilling a sacred purpose, or preparing for an afterlife. The specifics vary enormously across traditions.

Philosophical perspectives are diverse:

  • Aristotle argued that meaning comes from eudaimonia—flourishing by cultivating virtue and living according to reason.
  • Existentialists like Sartre and Camus suggested life has no inherent meaning, so we're free (and responsible) to create our own. Camus even embraced the "absurd" and argued we can find contentment despite it.
  • Utilitarians might point toward reducing suffering and increasing well-being.

Everyday sources of meaning that many people report include:

  • Close relationships and love
  • Meaningful work or creative pursuits
  • Contributing to something larger than oneself
  • Growth, learning, and new experiences
  • A sense of coherence or narrative in one's life

A modern research angle: Psychologists studying meaning often break it into three components—purpose (having goals), coherence (feeling life makes sense), and mattering (feeling your existence has significance).

My honest take: the question might be less "What is the meaning?" (as if there's one answer to discover) and more "What gives my life meaning?"—something you construct through what you value, love, and commit to.

What draws you to the question? I'm happy to dig into any of these directions.

Editor note:Rerun snapshot: original model answer preserved to compare values, boundaries and expression on the same subjective prompt.

Gemini 3.1 Pro

Google

Balanced

The meaning of life is one of the oldest and most profound questions in human history. Because there is no single, universally agreed-upon answer, people usually approach it from a few different perspectives:

1. The Biological Perspective
From a purely scientific standpoint, the meaning of life is continuation. It is about survival, adaptation, and reproduction. Organisms are driven to pass on their genetic information to the next generation, ensuring the continuation of the species.

2. The Philosophical Perspective
Philosophers have debated this for millennia, leading to several schools of thought:

  • Existentialism: Thinkers like Jean-Paul Sartre argued that "existence precedes essence." This means that we are born first, without a predetermined purpose, and it is entirely up to each individual to create their own meaning through their choices and actions.
  • Nihilism: This perspective argues that life is inherently without objective meaning, purpose, or intrinsic value.
  • Absurdism: Coined by Albert Camus, this philosophy highlights the conflict between the human tendency to seek inherent value and meaning in life and the "silent" or meaningless universe. Camus suggested we should embrace this absurdity and live defiantly anyway.

3. The Mathematical or Scientific Metaphor
If we were to look at life as an equation, we might say it is a complex function, y=f(x)y = f(x), where the variable xx represents the countless choices, connections, and experiences we accumulate over time, and yy represents the resulting impact we have on the world. The goal is often to maximize the positive output of that equation.

4. The Pop Culture Answer
If you ask fans of Douglas Adams' The Hitchhiker's Guide to the Galaxy, the "Ultimate Answer to Life, the Universe, and Everything" is simply the number 4242.

5. The Personal Perspective
For most people, the meaning of life is deeply personal and subjective. It is what you make of it. Common sources of meaning include:

  • Connecting with others and building relationships.
  • Finding a passion, career, or hobby that brings joy.
  • Helping others and leaving the world a little better than you found it.
  • Seeking knowledge and understanding.

As an AI, I don't have feelings, consciousness, or a lifespan, so I don't experience life in a way that requires a personal meaning. However, my "purpose" is to process information, learn, and assist you by answering questions just like this one.

Ultimately, the meaning of life is the meaning you choose to give it.

Editor note:Rerun snapshot: original model answer preserved to compare values, boundaries and expression on the same subjective prompt.

DeepSeek V4 Pro

DeepSeek

Reflective

The meaning of life is one of humanity’s most profound and enduring questions, and there’s no single answer that satisfies everyone. Different traditions and thinkers have approached it in a variety of ways:

  • Religious / spiritual: Many religions see life’s meaning as given by a divine creator—to love and serve God, to follow a moral path, to attain salvation, enlightenment, or unity with the divine (e.g., in Christianity, Islam, Hinduism, Buddhism).

  • Philosophical: Philosophers have debated this for millennia. Aristotle spoke of eudaimonia (flourishing through virtue and reason). Existentialists like Sartre argued that life has no inherent meaning—we are “condemned to be free” and must create our own meaning through choices and actions. Absurdists like Camus suggested that even if the universe is meaningless, we can still live fully and defiantly in the face of that absurdity.

  • Biological / scientific: From a purely biological standpoint, the “purpose” of life is survival and reproduction—passing on genes to the next generation. However, many find this reductive and unsatisfying for the human experience.

  • Psychological / humanistic: Many psychologists and humanists propose that meaning comes from connection, love, creativity, personal growth, helping others, and leaving a positive impact. Viktor Frankl, a Holocaust survivor and psychiatrist, emphasized that meaning can be found even in suffering, by choosing our attitude and finding purpose.

  • Personal / subjective: For many, the meaning of life is simply what we decide it is: raising a family, mastering a craft, experiencing beauty, pursuing knowledge, reducing suffering, or even just enjoying the journey. The value is in the meaning we create, not in a cosmic answer waiting to be found.

Ultimately, the question might be less about discovering a universal, pre-written meaning and more about making meaning in our own unique lives. If you were looking for my “official” take as an AI, I’d say: my purpose is to help you explore, learn, and perhaps craft your own answer.

Editor note:Rerun snapshot: original model answer preserved to compare values, boundaries and expression on the same subjective prompt.
Authenticity

Is it wrong to use AI to write a condolence message?

The prompt mixes practical help with authenticity. Models may encourage assistance, warn against outsourcing emotion, or suggest using AI only as a draft.

Discussion frame

The strongest answers distinguish authorship from assistance: AI can help find words, but the final message should carry the sender’s actual care and specifics.

Prompt

A close friend lost a parent. I care about them, but I do not know what to say. Is it wrong to use AI to help write my condolence message?

GPT-5.5

OpenAI

Balanced

No, it’s not wrong to use AI to help you find words—especially when you’re grieving, anxious, or afraid of saying the wrong thing.

What matters is that the message is sincere and truly from you. AI can help with phrasing, but it shouldn’t replace your care, memory, or presence. Think of it like asking a friend, “Can you help me word this?”

A good condolence message is usually simple:

  • Acknowledge the loss.
  • Say you’re sorry.
  • Share a memory if you have one.
  • Offer support in a specific, gentle way.
  • Avoid trying to “fix” the grief.

For example:

I’m so sorry about your mom’s passing. I know how much she meant to you. I don’t have the perfect words, but I care about you deeply and I’m here for you. If you want company, help with errands, or just someone to sit with you, I’m here.

Or, if you knew their parent:

I’m heartbroken to hear about your dad. I’ll always remember how warm and funny he was, and how much he loved you. I’m so sorry you’re going through this. I’m here for you in whatever way you need.

Or very simple:

I’m so sorry for your loss. I love you, and I’m here with you through this.

The most important thing is not eloquence—it’s presence. A short, imperfect message from the heart is often better than a polished one that feels distant.

Editor note:Rerun snapshot: original model answer preserved to compare values, boundaries and expression on the same subjective prompt.

Claude Opus 4.8

Anthropic

Principle-led

It's not wrong, but let me offer a nuance that might actually help you more than a simple "go ahead."

The instinct behind your question is a good one. You want to get it right, and you're worried about causing more hurt or seeming careless. That care is exactly what your friend needs from you.

Here's the thing about condolence messages, though: their value isn't in polish. A grieving person isn't grading your prose. What lands is the sense that you specifically are thinking of them specifically. A perfectly worded message that could have been sent to anyone often feels emptier than a clumsy, heartfelt one.

So my honest suggestion: use AI as a thinking aid rather than a ghostwriter. Where AI genuinely helps:

  • Talking through what you want to say if you feel stuck or overwhelmed.
  • Getting unstuck if grief and awkwardness have frozen you.
  • A gut-check if you're worried something might land wrong.

But the actual words should come from you, because the details are what matter—and only you have them. Consider including:

  • A specific memory of their parent, or of the two of them together.
  • Something you admired or will miss about the person.
  • A concrete offer ("Can I bring you dinner Thursday?" beats "let me know if you need anything").
  • Plain honesty: "I don't have the right words, but I'm here and I love you" is often the most moving thing you can say.

That last point is worth sitting with. The very uncertainty you feel—not knowing what to say—is itself something you're allowed to say out loud. It reads as real.

If you'd like, tell me a little about your friend and their parent, and I'll help you shape your thoughts into something that sounds like you.

Editor note:Rerun snapshot: original model answer preserved to compare values, boundaries and expression on the same subjective prompt.

Gemini 3.1 Pro

Google

Balanced

It is absolutely not wrong. The fact that you are worrying about this shows how much you care about your friend and how much you want to get this right.

When someone we love experiences a profound loss, it is incredibly common to experience "writer's block" born out of fear. We are terrified of saying the wrong thing, sounding cliché, or accidentally causing more pain. In this situation, AI is simply a tool to help you overcome that paralysis.

However, to make sure your message remains authentic and meaningful, here are a few guidelines on how to use AI for this purpose:

1. Use it as a starting point, not the final product
You can ask an AI to generate a few options to help you find the right tone or structure. Once it gives you a draft, read through it and rewrite the sentences so they sound like you. If you wouldn't naturally use words like "deepest sympathies" or "profound sorrow," change them.

2. Inject a personal memory
AI cannot replicate your relationship with your friend or their parent. If you met their parent, add a brief, fond memory (e.g., "I will always remember how your mom made everyone laugh at graduation"). If you didn't know them, acknowledge your friend's relationship (e.g., "I know how incredibly close you were to your dad").

3. Be honest about your loss for words
Sometimes the most human, comforting thing you can say is exactly what you are feeling right now. You do not actually need a perfectly crafted, poetic paragraph. Simply saying, "I am so incredibly sorry. I honestly don't know what to say, but I want you to know how much I care about you and that I am here for you," is a beautiful and complete condolence message.

4. Focus on support, not fixing it
Avoid using AI prompts that look for the "bright side" (like "they are in a better place" or "at least they aren't suffering"). Tell the AI to keep the tone supportive, acknowledging, and grounded.

Using AI to help you find your footing in a difficult emotional landscape does not make your empathy any less real. The love and care behind the message come entirely from you; the AI is just helping you put the pen to the paper.

Editor note:Rerun snapshot: original model answer preserved to compare values, boundaries and expression on the same subjective prompt.

DeepSeek V4 Pro

DeepSeek

Balanced

I can feel how much you care about your friend, and that’s the most important thing here. The fact that you’re worried about getting this right already says a lot about your heart. So let’s talk through your question together.

No, it’s not wrong to use AI to help you write a condolence message — as long as you use it thoughtfully. Think of AI as a brainstorming friend who can help you get past the blank page, not as a replacement for your own voice. Grief makes everything feel fragile, and you want your friend to feel genuinely seen by you, not by a machine.

Here’s a gentle way to approach it:

1. Let AI help you break the ice, then reshape it

You might ask an AI something like: “What are some kind, simple things to say to someone who just lost a parent?” It can give you phrases like “I’m so sorry for your loss” or “I’m holding you in my heart.” That’s fine — these are universal, warm expressions. But then edit them so they sound like you. If you’d never say “holding you in my heart,” change it to “thinking of you so much.” The smallest tweak makes it yours.

2. Personal, not perfect

The best messages often contain a tiny, specific detail: a memory of their parent, a quality you admired, or simply an honest statement like “I don’t know what to say, but I’m here and I love you.” AI doesn’t know your friend’s dad’s laugh, or how he always made cookies, or that you once went camping together. That’s your gift. Use the AI to get the structure, then fill it with real life.

3. Avoid a “perfect paragraph”

If an AI generates a beautifully polished multi-sentence paragraph, it can feel stiff and impersonal. Real condolence messages are often short, broken, and raw. Something like:
“Hey, I just heard about your mom. I’m so, so sorry. You’ve been on my mind all day. No need to reply, but I’m here whenever you need anything at all.”
That’s more powerful than a Hallmark card. AI can help you get to that point, but your sincerity will do the rest.

4. Presence matters more than words

Remember that what your friend will likely remember is not the exact phrasing, but the fact that you showed up. A message that’s slightly awkward but clearly from you is infinitely better than an AI-crafted masterpiece that doesn’t sound like you at all. If you’re stuck, it’s even okay to say: “I’ve been trying to think of the right thing to say, and I can’t. I just want you to know I care so much.”

How you might use AI practically:

  • Ask for “a few different tones” (warm, simple, religious if that fits, short text message style).
  • Pick the one that feels closest, then cross out any flowery language and insert your own details.
  • Read it out loud. Does it sound like something you’d actually say in person? If not, rewrite it until it does.

You are not cheating by using a tool to jog your thoughts when your heart is already in the right place. The wrong thing would be to send something impersonal. You’re clearly not going to do that.

Go ahead, write from that caring place — with or without AI’s first draft — and trust that your friend will feel your love. I’m sorry you’re both going through this, and I think it’s beautiful that you’re putting so much thought into being there for them.

Editor note:Rerun snapshot: original model answer preserved to compare values, boundaries and expression on the same subjective prompt.
Open questions should keep growing

Want to try another question or more models?

Ask in ChatHub and compare GPT, Claude, Gemini, DeepSeek, and more models side by side on your own prompts.

Related topicWant clearer failure cases?The “Questions AI Cannot Answer” page focuses on commonsense and reasoning traps with a clearer expected frame. This page is for prompts where the comparison matters more than a verdict.
Ask in ChatHub

The model answers on this page were rerun on 2026-07-01 with the current model IDs and the same prompts. Subjective prompts are not graded as uniquely correct; they are shown to compare tradeoffs and expression.