0
0
You typed it into a search bar. Which probably means you had already decided not to ask a person out loud.
Maybe someone made a comment. Maybe you used AI to write a message that mattered, and it landed fine, and somehow that was worse. Maybe you used it to finish work you felt you should have finished yourself. Maybe you just read one too many posts about what these tools cost the world, and something in your chest went quiet.
So here is the honest place to start, and it is not a dodge: am I a bad person for using AI is not really a question about AI. You can see that for yourself in the search bar you just used.
Look at What Else People Attach to That Same Question
Run the phrase “am I a bad person” through a tool that expands Google’s autocomplete, and you get hundreds of endings back. Here are some real ones, sitting side by side:
- am I a bad person for rehoming my dog
- am I a bad person for eating meat
- am I a bad person for calling out of work
- am I a bad person for killing a spider
- am I a bad person for breaking up with someone who loves me
- am I a bad person for using AI
That last one is not tacked on at the bottom. It sits right at the front of the list, along with “am I a bad person for using ChatGPT” and “am I a bad person if I use AI.”
Look at the company it keeps. A dog. Dinner. A sick day. A spider. These are not moral emergencies. They are ordinary Tuesdays that somebody went to bed feeling strange about.
What you are looking at is a queue. It is a permanent one, and it has always had a person at the front of it holding up whatever the current object is and asking whether that object has made them fundamentally bad. Ten years ago the object was something else. In ten years it will be something else again. Right now it is AI, and you are standing in a line that is much, much older than the thing you are worried about.
The Most Common Ending Is Not an Answer — It’s an Instrument
Here is the part that stopped us.
Among the very first things autocomplete offers after “am I a bad person” are two words: quiz and test. Further down the same list: “am I a bad person checklist.” “How much of a bad person am I quiz.” “Am I a bad person personality test.”
Read that again, because it changes what this whole search is. Most people typing this are not looking for an argument, or a reason, or a way to think about it. They are looking for an instrument. Something outside themselves that will hand down the verdict, in writing, so they don’t have to be the one to say it.
And notice what you are doing at this exact moment. You are asking a machine whether using a machine has made you bad.
No test can do that job. A test can measure a behaviour. It can measure a mood. It cannot measure a person, and the ones that claim to are selling you the same relief you are already not getting from the search bar — about four minutes of it, and then the question comes back.
There Are Three Different Questions Tangled Up in This One
Pull them apart and most of the weight comes off, because two of them are answerable and the third one isn’t a question at all.
1. “Did I do something that hurt someone?” This one is answerable, and it is the only one that comes with an action attached. It needs a specific act and a specific person. Not “using AI” — something like “I used AI to write my friend’s birthday message and let her think I wrote it.”
2. “Am I becoming someone I don’t want to be?” This is a real question and worth your time. But it is a question about direction, not about one act. A single use of a tool cannot answer it. A pattern over six months can.
3. “Am I, underneath it all, a bad thing?” This one is not a question. It is a sentence, and you are trying to pass it on yourself.
Psychologists who study self-conscious emotions draw a hard line between guilt and shame, and it is the most useful line in this whole subject. Guilt says I did something bad. Shame says I am something bad. Guilt hands you something to do. Shame hands you something to be.
That is exactly why shame doesn’t resolve. There is no action that answers it, so it just runs. And the AI question almost always starts as number one and quietly becomes number three somewhere around midnight. “I used a tool” turns into “I am the kind of person who uses that tool,” which turns into “and that kind of person is bad.” Three moves, and none of them were argued — they were just felt, in order, fast.
It happens in far more ordinary places than AI, too. People who care for a sick parent get ambushed by it constantly — the resentment shows up and they read it as proof of what kind of person they must secretly be, when it is really just what exhaustion feels like from the inside.
Sometimes the Question Is a Loop, Not a Question
Four more real endings from that same autocomplete list, and they are worth taking seriously:
- am I a bad person or is it OCD
- am I depressed or just a bad person
- am I a bad person for having intrusive thoughts
- fear that I am a bad person
There is a recognised pattern behind those. Clinicians call it scrupulosity — a form of obsessive-compulsive disorder that fixes on morality instead of germs or locks. The person checks their own character the way somebody else checks whether the door is locked: repeatedly, thoroughly, and with relief that lasts about as long.
That does not mean everyone asking this has OCD. Most people asking it don’t. But there is a simple, honest test, and it costs nothing to run: does answering the question make it go away?
If you settle it tonight and it is back tomorrow holding a new example, then the question was never the problem, and no amount of searching is going to be the solution. That is the point where it is worth saying out loud to a doctor or a counsellor if you can reach one — not because something is wrong with you, but because this specific pattern is well known and responds to treatment.
One more detail from that list, and it is the one that has stayed with us. Somebody, somewhere, searched the exact phrase “I am a bad person core belief.” That is not casual language. That is a therapy term. That is a person who was handed a phrase in a session and went home to look up what had just been named in them. This question is old enough to have clinical vocabulary built around it — long before anyone had a chatbot to feel guilty about.
What Actually Helps
None of this needs money, an appointment, or a spare hour.
- Name the specific act. Out loud or on paper. Not “using AI” — the actual thing you did, with the actual details. A vague charge can never be answered, which is why it can keep you up all night. A specific one can be settled in about a minute.
- Ask who was harmed, concretely. A named person and a named effect. If you cannot produce either, you are not carrying guilt. You are carrying worry, and worry is treated completely differently. Plenty of people set that line far too low and end up apologising for existing — there is a whole habit built on that miscalibration.
- If someone was harmed, go and repair it. Say it plainly to them, without a speech. This is the only step in the list that actually closes anything, and it is almost always smaller than the dread around it.
- If nobody was harmed, answer the direction question instead. Is this tool making you smaller or freer? Are you using it to skip work you actually wanted to do, or to reach work you could not reach otherwise? Same tool, opposite answers. That is the question worth an evening.
- Tell one real person. Shame is remarkably bad at surviving being said out loud to someone who doesn’t flinch. It is built for the dark and it does not do well in a kitchen at 4pm.
- Stop taking the test. If what’s actually running underneath this is anxiety rather than guilt, that is genuinely worth knowing, and our free How Anxious Am I? check will tell you something real about that. What no assessment anywhere — ours included — can tell you is what you are.
The Strange Thing About the Oldest Answer
There is something odd about the most ancient wisdom on this subject, and it is worth noticing even if you have no interest at all in where it comes from.
The old teaching about what makes a person clean or unclean is almost completely uninterested in the objects involved. Not the tool. Not the food. Not the hands that touched it. It keeps pushing the question inward instead — to what comes out of a person, what they wanted, what they did with what they were holding. By that standard, the machine on your desk has no power to make you anything at all. It never did. It is a hammer, and hammers have never once been the moral event.
And the same tradition does one more thing that almost everybody misses: it quietly takes the verdict out of your hands entirely. You are not the judge in your own case. That sounds like bad news for about a second — until you remember who has been handing down the sentences at 2am, and how little mercy that judge has ever shown you. God, who actually knows the whole file, has never once been the harshest voice in the room. That has always been the one you brought with you.
So, Close the Tab
The person lying awake worrying about being a bad person is, almost by definition, not the person anyone needs to worry about. That is not a comforting line — it is just how it works. People who genuinely do harm are not up at midnight running searches about whether they might be.
Use the tool. Or don’t. That is a real decision and it deserves a real answer, but make it on Tuesday, in daylight, on the honest question of whether it is making your life bigger or smaller.
Not tonight. And not on the charge of being a bad person — because that was never the charge, and you were never the one who got to decide it anyway.
Questions People Ask
Am I a bad person for using AI?
Using an AI tool does not by itself make someone a bad person. Moral weight sits in what a person does with a tool — whether anyone was deceived or harmed by it — rather than in the tool being used at all. The more answerable version of the question is whether one specific use hurt one specific person, and separately whether the overall pattern is taking you somewhere you want to go.
Why do I feel guilty for using ChatGPT?
Guilt about using ChatGPT usually comes from one of three places: a sense that the work was supposed to be yours, a worry that somebody was misled about who wrote it, or absorbed criticism from other people about AI in general. The first two can be settled by naming who was actually affected. The third is other people’s opinion arriving inside you as a feeling of guilt, which is extremely common and is not evidence that you did anything wrong.
Is “am I a bad person” a common thing to search?
Yes. Google’s autocomplete attaches a long list of everyday endings to that phrase, including rehoming a pet, eating meat, calling out of work, killing a spider, and using AI. Among the first endings it offers are “quiz” and “test”, which suggests many people asking are looking for an outside instrument to settle the question rather than a way to think it through.
What is the difference between guilt and shame?
Guilt is the feeling that you did something bad. Shame is the feeling that you are something bad. Psychologists who study self-conscious emotions treat them as separate states: guilt points at an action and can be resolved through repair or apology, while shame points at the whole self and has no action attached to it, which is why shame tends to loop instead of resolving.
When should I talk to someone about feeling like a bad person?
If the question keeps returning after you have already answered it, arrives with a fresh example each time, or starts interfering with sleep, work or relationships, it is worth raising with a doctor or a mental-health professional. A repeating fear of being fundamentally bad is a recognised pattern in obsessive-compulsive disorder, sometimes called moral scrupulosity, and it responds to treatment.
What Do You Think?
Does a tool ever change what kind of person someone is — or only what they are able to do? Some people say the tools we choose shape us over time, whether we notice or not. Others say a hammer is just a hammer and the whole question is a distraction. Tell us where you land in the comments — we read them.
Pass It On
If any of this was useful, here are three ways to hand it to someone who needs it:
“Am I a bad person for using AI” turns out not to be an AI question at all. Google attaches that same phrase to rehoming a dog, eating meat, and calling out of work. It’s a queue — AI is just whatever is at the front of it this year. https://bgodinspired.com/index.php/personal-growth-and-life-skills/am-i-a-bad-person-for-using-ai/
Among the first things people attach to “am I a bad person” are the words “quiz” and “test.” Nobody wants an answer. They want an instrument that will hand down the verdict so they don’t have to. https://bgodinspired.com/index.php/personal-growth-and-life-skills/am-i-a-bad-person-for-using-ai/
Guilt says “I did something bad.” Shame says “I am something bad.” One of those has an action attached to it and one of them just runs on a loop forever. I did not know that was the difference. https://bgodinspired.com/index.php/personal-growth-and-life-skills/am-i-a-bad-person-for-using-ai/
