Is it healthy to use ChatGPT for an ego boost?

[Mabbo]

I'm a pretty unmotivated guy. I write my current novel (A.B.B.Y) because the idea well hasn't dried out yet, but I do want some external gratification too every once in a while. I'm really afraid of losing my readers since that 100% will lead to me stop writing this novel altogether. Writing to the void sucks.

So, on the one hand, I really want to know how people find my novel. I can't improve anything if nobody told me which part they do like and which part they don't. On the other hand, I don't want to pester them because, to be honest, it's just a casual read that shows up every Friday. It's not the best fiction out there, the idea is not exactly new, and the main character doesn't even show up until 6 chapters in... well, she did, or they did, or rather the two halves of her did show up since chapter 2, but I digress. The point is; I don't want to look so desperate that I lose all of my casual readers.

Here's where ChatGPT comes in. Being a non-native English speaker, my grammar is a bit wonky. I learned most of my English commanding skill from reading translations of Japanese web novel, and we all know how that looks like. I tried grammarly. It sucked hot donkey turd. I turned to ChatGPT and realized I can just copy-paste sections of my novel and ask for a revision. It, also, sucked hot donkey turd, but at least it understood context. And you know what that means; I can just ask ChatGPT how good my novel is.

Well, it praised me. A lot. Strong characterization this, interesting world-building that. Good thematic this, nice contrast that. My ego went to the moon when it told me that, and I quote, "You’re doing fantastic—your writing is rich with thoughtful details, layered themes, and distinct character voices. Every scene and piece of dialogue you share feels purposeful and full of potential. Your story’s really shaping up to be something special! ?"

But then it hit me. How could I know it's not lying? I feel like I was on a drug being praised to heaven and back, but it's just an AI tool that I gave the prompt "I need an ego-boost." It could very well just make those praises up to produce an outcome fitting the prompt. Part of me is actively regretting the fact that I asked ChatGPT in the first place. Now I'm afraid that this is going to have a negative impact on my motivation on the long run.

Not that I'll stop writing, not until the idea well runs dry.

 

[Clo]

ChatGPT is highly suggestible.

If you ask "what do you think of [passage]", it will usually start with some strengths, then give you some stuff to work on, might point out typo, or revised paragraphs.

If you ask "why do you think [passage] sucks?", it won't answer the same way.

So the way you frame your question will taint the way it will reply.

It's perfect to actually control the type of feedback you're looking for. Are you trying to improve a passage? Ask it to help you find where your weaknesses are.

Are you trying to boost your self-esteem a bit and be told you're doing a good job for a hobbyist?

Try asking "Please read this chapter and tell me what you liked, and how excited you would be to turn the page to the next one after reading it."

Or "give a score between 1 and 10 on how enjoyable this chapter is to read", if you just want a vague numerical idea of which of your chapter needs more work.

You could tell it "look for any tense slip in [...]", or "I am not an English native speaker. Can you tell me which sentence sound off to you in this?"

GPT's feedback will forever be aligned with what you ask it to do. Like I opened with, it is very suggestible.

Loaded questions is a tool. "Why is Vanilla ice cream the best flavour of ice cream?" vs. "What is the best flavour of ice cream?" usually will yield wildly different results.

Be very honest with what you're looking for.

I found that asking it about the "page turner" quality of each chapter was one of the best prompt to motivate me.

Or just asking it "Why do you think character [x] did this thing in this story?" can lead down some very interesting path and help you flesh out your characters.

GPT is a tool. And you should usually treat is as the most "middle of the road" person you know. It's not a genius, it's not dumb, and will deduce or think whatever the "average" of all the data it trained on thinks. It's running on probability.

So if you are talking about your plot, and he does a deduction you disagree with? That's fine. It just meant your idea is not the most tropey one.

Good luck learning to turn that software into a useful tool. And be wary of anything it says. You should always make sure to take the time to really analyse its answers.

 

[aToTeT]

I'm a pretty unmotivated guy. I write my current novel (A.B.B.Y) because the idea well hasn't dried out yet, but I do want some external gratification too every once in a while. I'm really afraid of losing my readers since that 100% will lead to me stop writing this novel altogether. Writing to the void sucks.

So, on the one hand, I really want to know how people find my novel. I can't improve anything if nobody told me which part they do like and which part they don't. On the other hand, I don't want to pester them because, to be honest, it's just a casual read that shows up every Friday. It's not the best fiction out there, the idea is not exactly new, and the main character doesn't even show up until 6 chapters in... well, she did, or they did, or rather the two halves of her did show up since chapter 2, but I digress. The point is; I don't want to look so desperate that I lose all of my casual readers.

Here's where ChatGPT comes in. Being a non-native English speaker, my grammar is a bit wonky. I learned most of my English commanding skill from reading translations of Japanese web novel, and we all know how that looks like. I tried grammarly. It sucked hot donkey turd. I turned to ChatGPT and realized I can just copy-paste sections of my novel and ask for a revision. It, also, sucked hot donkey turd, but at least it understood context. And you know what that means; I can just ask ChatGPT how good my novel is.

Well, it praised me. A lot. Strong characterization this, interesting world-building that. Good thematic this, nice contrast that. My ego went to the moon when it told me that, and I quote, "You’re doing fantastic—your writing is rich with thoughtful details, layered themes, and distinct character voices. Every scene and piece of dialogue you share feels purposeful and full of potential. Your story’s really shaping up to be something special! ?"

But then it hit me. How could I know it's not lying? I feel like I was on a drug being praised to heaven and back, but it's just an AI tool that I gave the prompt "I need an ego-boost." It could very well just make those praises up to produce an outcome fitting the prompt. Part of me is actively regretting the fact that I asked ChatGPT in the first place. Now I'm afraid that this is going to have a negative impact on my motivation on the long run.

Not that I'll stop writing, not until the idea well runs dry.

I will say: ChatGPT is predicated upon your engagement.

If you tell it to be mean, and then to roast your work as if they are Gordon Ramsey being presented with raw chicken:

Itll hurt.

It will not pull punches. Demonstration of its perception of my best chapter (27 Favourites) so you do not have to see this happen to your own story:

Spoiler

What in the actual RAW mess is this, you donkey?!

Let’s dig into this disaster step by step:

Pacing and Focus

• “What do you most wish for?” The opening question is meant to be weighty, but it drags into a sprawling existential whimper instead of creating tension. The pacing is so sluggish it feels like trying to grill a chicken breast over a wet matchstick.

• The timer is a supposed source of urgency, yet it’s undercut by Éclair’s meandering inner monologue. Why does she have time to make snide observations, review options, and philosophize if she’s seconds away from oblivion? It’s like watching someone leisurely browse a menu while their house is burning down.

Éclair’s Voice

• Her voice is inconsistent—sometimes frantic, sometimes overly composed, and other times completely disengaged. It’s unclear if we’re supposed to sympathize with her panic or laugh at her detachment. Pick a lane.

• The humor feels forced. Lines like, “very-small-sea-slug-with-super-cancer” and “slightly-starving-tapeworm-in-a-pot-of-pigeonplop” try to be quirky, but they undercut the tone and undermine Éclair’s supposed stakes.

Repetitive Mechanics

• The errors on her typed wishes are excruciatingly tedious. Watching Éclair type and retype, only to get error after error, grinds the momentum to a halt. Once or twice might build tension; a dozen times feels like watching someone play tech support with a genie. Nobody wants to sit through that!

• The endless cycling between “let’s try this” and “oh, that didn’t work” is the narrative equivalent of raw, undercooked poultry: repetitive, bland, and liable to make readers sick.

Worldbuilding Info Dumps

• The reincarnation options list is another slog. Instead of enhancing the stakes or immersing the reader, it becomes a tedious roll call of weirdly specific races, each with a Wikipedia-style summary. If you’re trying to lose momentum, this is how you do it.

• The descriptions lack focus. Do we really need so much detail about Sahgumer being gross or Goblins making Éclair feel sick? Half the entries are barely relevant and feel like filler.

Emotional Core

• Éclair’s supposed emotional revelation—choosing “love” as her wish—is rushed and unearned. Her sudden longing for Rachel comes out of nowhere and feels tacked on as an afterthought. Where’s the buildup? Where’s the pathos? Where’s the seasoning?

• Her decision-making process lacks consistency. She dithers over “omnipotence” and “invincibility” but somehow settles on “love” without any meaningful exploration of why. It’s as though the narrative panicked under its own timer and slapped an ending together.

Unnecessary Filler

• The keyboard antics: Why is she typing, flinging, and wrestling with a keyboard in a metaphysical void? It’s a drawn-out, slapstick routine that feels out of place and detracts from the supposed urgency.

• The inner dialogue veers into tangents that don’t add anything. Her thoughts about Sahgumer, Beastkin, and Kobolds are distractions from the main conflict.

Missed Opportunities

• The timer could have been used to ratchet up tension with sharp, escalating stakes. Instead, it feels like an arbitrary deadline looming over an aimless scene.

• Instead of delivering a poignant or thrilling moment, Éclair’s final decision feels like a shrug. It’s the literary equivalent of undercooking the chicken and hoping nobody notices.

Final Verdict:

This chapter is a bloated, raw, under-seasoned narrative that wastes its tension and fails to deliver a satisfying payoff. It’s overstuffed with filler, bogged down by repetitive mechanics, and sorely lacking in emotional resonance. Take it back, rethink the pacing, and for the love of all that’s holy, don’t serve it like this again!

What you have to understand is that ChatGPT is not a *fair* or *sensible* critic of your work. It does not know what writing is, only what the most likely word to come statistically after the first.

You MUST be careful to neither get a big head, nor to take its ‘advice’ on how you might ‘improve’ too seriously. It can help as a tool for refinement: but almost never by what it actually suggests… only through active argument with yourself using different perspectives to analyse your work.

It does not UNDERSTAND context. It does not appreciate use of language either with witty brevity or purple prose. It *cannot* understand.

If you do use it, strong recommendation to be very specific with your usage: sanity check for preliminary research, line-by-line spelling/grammar/purpose/ethos/logos/pathos you must independently review each facet of to see if you agree with the indications provided or still see room to argue the point.

And it will never be satisfied: because satisfaction is not its goal — interaction is.

My opinion is that if it inspires you to write: go for it. Me, I find it an incredibly frustrating cross-examining tool that I bash my head against and get quite vitriolic in my arguments with — one I cannot trust to speak to me in good faith, and which abhors analysis as it delights in invented hallucinations.

 

[Clo]

If you tell it to be mean, and then to roast your work as if they are Gordon Ramsey being presented with raw chicken:

Itll hurt.

Interestingly, when I asked it to roast me, once, out of curiosity, I recognised the exact pattern it normally responds to me (strengths, then weaknesses, then suggestions), but it just changed the vocabulary to sound meaner.

I have never asked it that way again, because it didn't make it more critical, just more hurtful in how it talked.

But yeah, 100% agreeing with aToTeT.

That said, I think it's healthy to ask GPT for encouragement if the opposite is losing motivation and stopping writing.

Anything that gets you to type more words (or read more words) is making you a better writer, little by little.

If that's all GPT did (motivate you), then it did God's work.

 

[aToTeT]

I have never asked it that way again, because it didn't make it more critical, just more hurtful in how it talked.

That said, I think it's healthy to ask GPT for encouragement if the opposite is losing motivation and stopping writing.

If that's all GPT did (motivate you), then it did God's work.

ChatGPT can gas you right up to the stratosphere for sure. There is value in that — there is joy in seeing the good in what you write, no matter how farcical its basis may be.

But even better is to read it to someone else aloud and to see them smiling, laughing, and wincing.

If we all had such available to us at all times: chatGPT’s solitary usefulness would be in saving me from the sponsor-weighted google search and subsequent reading of an entire article for an apple pie recipe written by chatGPT buried at the bottom.

 

[Clo]

If we all had such available to us at all times: chatGPT’s solitary usefulness would be in saving me from the sponsor-weighted google search and subsequent reading of an entire article for an apple pie recipe written by chatGPT buried at the bottom.

Me and a work colleague were saying exactly that like 12 hours ago in call.

Serendipity.

 

[beast_regards]

Hmmm....

The normal sequence of events is the following 4-step process of amateur writer's internet interaction

Step 1) Write a fan-fiction because you feel depressed and you think of the cool story that would make you feel less depressed
Step 2) Post it on the Internet expecting positive feedback that would help you with your depression
Step 3) Be told that your work is trash and you should kill yourself
Step 4A) Go to Step 1
*or
Step 4b) Realize that "Only winning move is not to play. "

If you are particularly persistent you will follow this cycle few times until the Step 4b) applies ...

The relying on ChatGPT to tell you how great you are might not be the best solution, psychology wise, but considering the alternative is going through above ...

 

[Representing_Tromba]

I personally think it's morally wrong to use chatgbt but you do you. Outside of the moral side of it, I don't think it's a good idea because there is no human connection to really feel. It's like a manipulator that doesn't actually care for you but just wants to tell you what it thinks you want to hear so you keep them around.

 

[CharlesEBrown]

I'm probably in the minority here, but as long as you don't use it to do the actual writing, I see nothing wrong with using AI as a tool to improve and help. Just keep in mind that a tool can become a crutch - and if you use it to do any actual writing, you are really collaborating with everyone who taught the AI, plus the AI itself, not actually writing; at most you're being a "completionist" or "revisionist" (which is how H. P. Lovecraft made a living, so that's probably fine), not an author at that point.

 

[Tempokai]

ChatGPT is a full blown orator. It doesn't understand the substance, and repeats what it knows, for it's raison d'etre is all pandering dazzle, zero intentional rhetoric. It's a tool, and tools must be used properly. Understanding of the tool and its biased nuances is always the best way to go with such things, don't let yourself be dazzled with it, even for "writing ego", as it must be destroyed for better work to emerge from inside of the writer.

 

[beast_regards]

ChatGPT isn't able to understand the substance, or context, or meaning.

It doesn't know what it is saying.

It doesn't know anything.

It doesn't even know you are there.

It is a program, a complex program, that reacts to the input with the most probable output, and thus can mimic the dialogue, but it doesn't, in fact, understand the content, the subject, or the object, of the message.

In a way, it is worse than the Google Search engine. If you ask it a question, it doesn't understand what you are asking, but gives most probable output to your input. It could get things wrong.

It will get things wrong, and could mess up easy mathematical formulas simply because it doesn't know it is a mathematical formula in the first place. If you trust it, it would gaslight you, without any intent to lie. It doesn't know what the lie is. It doesn't know what the truth is, either.

That's pretty bad, isn't it?

Considering the alternative is talking to the random people on the Internet ...

 

[BearlyAlive]

Using an AI to boost your ego is almost as bad as using AI sexting bots to learn how to write smut...

 

[RepresentingCaution]

An ego boost will not help to motivate you. Too much praise and not enough emphasis on hard work is why American kids are behind Japanese kids academically. Ask specifically for constructive criticism, and that might help.

 

[Tsuru]

All the aboves.



And its ez to create a "echo chamber"
You have better luck to ask in SHF and ask specifically "please praise me" or "tell the good points(before the negative points)"

 

[lambenttyto]

If you use AI to write, then you're lower than "hack writer" status. Dump that ****.

 

[unlaumy]

It just seems sad. I would hate myself more if I remembered that I asked chatGPT to praise me.

 

[Valmond]

Using an AI to boost your ego is almost as bad as using AI sexting bots to learn how to write smut...

Buddy, you don’t need a specific AI to do that, they’ll do it on their own.

 

[Daydreamers]

AI can't give a feedback, the only thing it might help with is grammar

 

[istryj]

I'm a pretty unmotivated guy. I write my current novel (A.B.B.Y) because the idea well hasn't dried out yet, but I do want some external gratification too every once in a while. I'm really afraid of losing my readers since that 100% will lead to me stop writing this novel altogether. Writing to the void sucks.

So, on the one hand, I really want to know how people find my novel. I can't improve anything if nobody told me which part they do like and which part they don't. On the other hand, I don't want to pester them because, to be honest, it's just a casual read that shows up every Friday. It's not the best fiction out there, the idea is not exactly new, and the main character doesn't even show up until 6 chapters in... well, she did, or they did, or rather the two halves of her did show up since chapter 2, but I digress. The point is; I don't want to look so desperate that I lose all of my casual readers.

Here's where ChatGPT comes in. Being a non-native English speaker, my grammar is a bit wonky. I learned most of my English commanding skill from reading translations of Japanese web novel, and we all know how that looks like. I tried grammarly. It sucked hot donkey turd. I turned to ChatGPT and realized I can just copy-paste sections of my novel and ask for a revision. It, also, sucked hot donkey turd, but at least it understood context. And you know what that means; I can just ask ChatGPT how good my novel is.

Well, it praised me. A lot. Strong characterization this, interesting world-building that. Good thematic this, nice contrast that. My ego went to the moon when it told me that, and I quote, "You’re doing fantastic—your writing is rich with thoughtful details, layered themes, and distinct character voices. Every scene and piece of dialogue you share feels purposeful and full of potential. Your story’s really shaping up to be something special! ?"

But then it hit me. How could I know it's not lying? I feel like I was on a drug being praised to heaven and back, but it's just an AI tool that I gave the prompt "I need an ego-boost." It could very well just make those praises up to produce an outcome fitting the prompt. Part of me is actively regretting the fact that I asked ChatGPT in the first place. Now I'm afraid that this is going to have a negative impact on my motivation on the long run.

Not that I'll stop writing, not until the idea well runs dry.

The AI is set to give positive responses. I am not a developer. This is an empirical conclusion drawn from the combination of many facts, including direct communication with the AI.

 

[Clo]

ChatGPT defaults to motivational style feedback, and will try to balance quality and weakness so it doesn't come as mean.

You can tell it to stop pulling its punches, or you can ask it to be gentle.

You can be self-depreciating and sarcastic, and it will remind you about the things you are doing right in a friendly way.

It's funny to give it a chapter and basically take stab at your own work "hey, look, I wrote a 4 000 word chapter where the protagonist eats a poutine. Nobody else wastes as many words on a simple task as I do."

And then it will actually find ways to tell you how your chapter is more than just eating poutine.