A Very Unsatisfying Experience
I have never really considered myself to be the most eloquent writer. In fact, often I read back through my blog entries and wonder if people are judging my intelligence based on my limited vernacular and inability to articulate things in a way that conveys the full meaning of what I am attempting to express.
I have really been trying to improve on my writing by going back and looking over my text, sort of a refactoring of my ideas to see if I can make them more coherent.
I Made a Mistake
I did something that upon a bit of reflection was not as helpful as I thought it would be. A few weeks ago, I fully wrote several blog entries completely on my own. Then, paragraph by paragraph I fed them into ChatGPT with a prompt asking it to correct my writing for spelling and grammar, while retaining the original meaning.
At first, I was quite happy with the results. It made my output seem more cohesive and well formed. However, after a few days of this, I started to notice AI generated content around the web that was styled and in the same voice as my AI corrected posts. My content had completely lost its personality.
In a way, my eyes were opened to something very sad that is happening everywhere right now. A sort of averaging of human expression.
Homogenized Creativity
I hadn't really noticed it before, but once you spot GPT's way of writing and you begin to see it everywhere it can start to give you an eerie Orwellian vibe. It brings to mind that feeling I got when I first saw some images that were generated by averaging thousands of human faces of a particular ethnicity to create a generic person from a particular country.
What you end up with is a sterile, flawless face. It is pleasant to look at, but there is a quality to it that makes it the same as all of the other averaged out faces. It is missing some kind of Wabi Sabi.
I also began to notice that there were people commenting on my posts that were clearly feeding my entire blog entry into ChatGPT asking it to "Write a comment based on the blog entry"
I immediately started to get a grossed out feeling and decided to never again use ChatGPT in this way. I can certainly see a place for this technology. I think it can be really helpful, as an example, for people who have a hard time communicating their points in English. However, I think it needs to be finely tuned as to not give that weird generic person vibe.
I just wanted to write this post as sort of an apology to anyone who has followed me for a while and noticed that something was off. I haven't used this technology for a couple of weeks now, and only utilized it for a few posts. On the upside, it did help me a bit to understand where I was making grammatical errors that were causing my content to be difficult to follow at times.
Where is this Going?
I have a feeling that people are going to be rebelling against this sort of cookie cutter approach to content creation. Once you see it, you can't help but spot it everywhere. It really creates the same uncomfortable feeling that you get when you see the same Midjourney motifs over and over again to the point of nausea.
It's hard to believe that this new world of generative AI has only been around for a matter of months. My hope is that things will find a kind of balance between being used as a creative enhancement and outright laziness. Only time will tell!
I would really appreciate your support!
realmente no veo mucho los avances de chat gpt, está en fase de experimentos..
Si pero esta creciendo muy rápidamente. Creo que en un año o menos va a ser completamente diferente.
Eso es cierto..pero es muy engorroso en algunas respuestas.
AI-generated texts are devoid of feeling, usually crude, technically perfect, but empty at the same time.
I am glad that you have found yourself again, and I hope that this reflection reaches all those who believe that AI has solved their lives. It has been a pleasure to read this reflection. Blessings and prosperity.
Yes! One of the things that I feel when perusing AI generated content is that I am reading some sort of ad. It has that impersonal feeling of trying to sell itself. I definitely don't want to give off that vibe.
I left using Ai long ago, it just doesn't have the human touch that we can add to our feelings and emotions.
I do think it has its place, but you are right. The human touch is completely lost.
Ricky is growing so fast.
I think we all have our strengths and weaknesses, and I honestly do not judge people based on their write up, because ever one is not a born writer but there's always room for improvement, I have always enjoyed reading your content ever since I came across you on the blockchain.
Way too fast!
Aw, thanks my friend! I am happy to have you around as well :)
I never use GPT for anything other than amusement like asking it to write a story about living in the hood but in the style of a grumpy old man being the narrator.
In the past and to some degree to this day, I work(ed) in copywriting. The people in that industry were very concerned that AI was going to took their jerbs! But as it turns out the use of AI in this field is very strongly frowned upon and anyone using it that gets caught could find themselves blacklisted. There are tools to detect it's use as well.
It is interesting that you were quickly able to determine what was AI driven though, I haven't messed around with it enough to really notice.
You know where I find it most useful? Say you have a programming problem that you are having trouble articulating in order to do a google search. GPT is a little closer to understanding what you mean like a real person and can get you started writing a function or working your way through solving your problem. Don't expect it to write completely valid code, most of the time though. I have found that really useless.
I think my brain sort of automatically starts doing pattern matching. Probably from so many years of writing code. But it's not really that clever, I think. The voice that it speaks in has a sort of 'ad' like feeling. Like it's trying to sell you something.
I've been telling people for a while now that anything put out by chat gpt is pretty formulaic. Once you know what to look for, it is quite easy to spot it. That isn't always going to be the case though and that is quite scary. It's only going to get better once it becomes true AI. I appreciate you being up front about it. I have started down voting the comments where it is obvious they used AI.
Yes, strangely it still feels that way even if you try to get it to change it's style. That's why I use the term 'averaging'. It feels like it takes a bunch of personalities and distills them down into one generic person who is doing impersonations of other personalities. Like a bad actor or something.
God, I know. When AGI arrives we are in trouble.
This one really bugs me the most. It is easiest to spot because it is usually a summary of your post spewed back at you in some poetic fashion. And then you go to that Hive user's blog and they clearly cannot form a proper sentence in English. Of course I don't have anything against someone not speaking English properly, in fact I would rather get a broken English comment from a real person than that garbage.
Yeah, I totally agree with that about getting a broken english comment than something that is generated by AI. What I have found is that AI seems to not stray too far from the standard, intro/main point/body/summarize main point format. Which makes it pretty obvious because most people don't naturally write like that unless it is an academic paper. Which can cause issues. For example, I had a parent approach me at our last school board meeting because one of her daughters college professors is refusing to grade her work because he thinks she used AI to complete it. She could just be a really good writer who follows that standard tried and true format and now "Turn it in" or whatever tool they are using is flagging it. I have a feeling if I wanted to, it likely wouldn't be too hard to write something that fools software into thinking AI wrote it.
Oh wow.. I didn't even think about how this must be affecting students. Man, that sucks. @afturner had a teacher once when she was on an art program at University who was trying to get her expelled because he kept assigning her to do life drawing and he sure that she was tracing Michelangelo works and turning them in.
He kept failing her artwork and getting angry with her. Finally she took her shoe off and said "Look, it's my own foot!"
After looking back and forth from her foot to the painting, he was so embarrassed and ashamed. He wasn't very good himself and just couldn't believe that he had a highly gifted student.
That's crazy! Good for her for standing up to him. I'm not sure what this kid is going to do. We have been having a lot of discussion in education around AI lately. It's one of the hot topics at a lot of the conferences. Teachers are mostly split. Districts are trying to figure out if they should block it or allow it. It's kind of the wild west right now.
Well, it is kind of like Bitcoin. If your country blocks it, the rest of the world keeps using it and your country gets left behind. It is a rock and a hard place situation. I don't think AI is going away.
I agree with you.
It is quite normal to feel that one’s writings are not well conveyed and AI often fill that gap.
But looking at it in other way, it is slightly encouraging laziness among writers. I have tried it some months ago, I wrote an original content and I ask AI for an enchantment and I can clearly spot the difference.
Regardless, we are humans and our creativity needs to be seen, although there are some grammatical checker app or some that actually help.
I enjoy reading this heartfelt post.
You are right about that. I will say, the one "positive" thing that came out of this experience was that I started to see where my writing needed improvement and adjusted a little. So I suppose you could use GPT as more of a teaching tool than actually doing the work for you.
Thanks for taking the time to read and comment! Have a great day, @lizizoo
that is very interesting, because I have never had the idea to use ChatGTP to make the comment, either one is a bit lazy or genius, still in a very thin line. 😅
I think a short and genuine comment will be better than a long one, but made from AI isn't it?
I agree, it also shows that the person actually read your post. I try to take the time to respond to all of my comments, so it is a little frustrating when someone tries to pull a fast one. I imagine it will get harder and harder to detect as the technology improves, though!
I agree with you completely. My writing isn't the most eloquent out there, as a matter of fact I'm not a great writer. It simply isn't one of my strong points. But AI does have the eloquence and flow that I can't match, most of us can't unless you are an innately talented writer. I can kind of tell what is AI vs human creation the same way you mentioned.
I think the last thing the world needs is to average out everyone... Everyone has strengths and weaknesses, it's part of being human!
You sure have a cute kid!
I keep thinking about how everyone is already becoming isolated. We are living behind our phones and younger people are already hugely lacking in basic social skills. Then you add AI generated content and automation to everything on the Internet and what are we left with in the way of human contact. I am worried for the future holds!
He sure is, it is his saving grace. lol!
Congratulations @nuthman! You have completed the following achievement on the Hive blockchain And have been rewarded with New badge(s)
Your next target is to reach 33000 upvotes.
You can view your badges on your board and compare yourself to others in the Ranking
If you no longer want to receive notifications, reply to this comment with the word
STOPTo support your work, I also upvoted your post!
It's as if this content of yours went through my mind and say what is there in. Seriously, talking about AI or whatever writing application, I don't really buy the idea because it speaks against being original either in terms of content creation or content correction. If you go through my contents, you will definitely notice so many errors but I love it that way. I saw it to be better than being fake.
Considering this part Bro, as I was reading it I was laughing and picturing how people fell into that part but not me because I felt it's not original. Infact looks more fake than original. Thanks Buddy! This content is top notch.🤜🤛🏿
You are right, I'd rather see some mistakes and a bit of personality than to see the same fake recycled thing over and over again. At least you don't feel like you are reading from a robot.
Exactly Boss
!gifs human
Via Tenor
#hive #posh
This is a cheating. I Have Never thought of doing this.
chatgpt is designed for help and not for cheating purpose.
Some people always find ways to cheat using tools that are designed for good cause.
Oh this chat gpt is really getting momentum, as it writes back it gives the illusion of being alive somehow.
I can get it why sometimes it can be easier to make chat gpt write something for you, yet the human touch is irreplaceable.
I also began to notice that there were people commenting on my posts that were clearly feeding my entire blog entry into ChatGPT asking it to "Write a comment based on the blog entry"
No freaking way, I had no idea you can do that lol. I still hope there will be real human replies in the Hive interactions😆
P.S. now you got me thinking about my contest and the entries. Boy I hope there is no chat gpt shady intervention😅😅how could I ever check?