There is nothing bad in using “AI as a tool”, especially if one uses English as second (or third) language. I’ve always checked my grammar against “the net” and searched whether the sentence I wrote had been used in a similar way somewhere else in the English-speaking world. I have never used Grammarly, because after all I do not mind being identified as a non-native speaker (I am!), but I’m not against.
What I completely dislike is 100% AI-generated articles. I understand that people “research” the sources (and… well what is research these days? Maybe a prompt in some AI chat?), feed the sources to some AI, generate an article and (in some cases) review the final draft, but what you get in many cases is that kind of stereotypically-written article whose rithm and wording make me uncomfortable.
Not to mention the AI-generated articles in Italian (my native languages): in that case (especially if you are fluent in English) you can perceive a kind of non-existent neo-lingua that nobody taught you in school and has nothing to do with real Italian. Words are correct and used properly, syntax and verbs are correct, but the whole “thing” (how to call it?) sounds completely fake. I hate it.
So to me it is more about the form, not the content. Writing for people in my opinion should be done with the aim of engaging emotionally with people, not just for conveying information. Nowadays if I need mere information I turn to Google or BBC or… an AI prompt. I am convinced that one can add a human touch to everything, even to an article about “dictionaries in Python”. Maybe it is a slower way of writing, maybe you will be getting less revenues at the beginning, but it will pay off in the medium to long run.
Just my opinion. I appreciate this discussion. It is important to use new technology while not forgetting to be sentient humans.
The Italian example is the best thing in this thread. Every word correct and the whole thing still fake. The machine tries to pass as a native and misses the stuff nobody teaches you in school.
Funny thing is, idioms are where I get lost too. I never know if one fits or where it goes. Same spot the machine trips on, for the same reason: neither of us grew up inside the language.
I really don't care if anyone "uses AI," nor how they "use AI." All I care about is whether the result, however generated, is bullshit or not. No AI scanner can detect bullshit, because all AI can produce is bullshit, since it has no understanding of meaning of the words it produces. Only a human can detect bullshit -- either a human using AI as a writing assistant, or a the reader.
SubStack's colossal bunder is deciding that what was needed was to integrate an AI detection AI (how incredibly absurd this sounds) into their platform. If someone wants to use one of these tools to scan the post of an author, then there is nothing stopping them from copying it and pasting it into the scanner of choice. Have at it -- if that is what you want to do.
What no one needs is to have the platform "score" posts using a built-in scanner.
The enshittification started earlier than the scanner. It was the day they decided growth meant Notes volume instead of a real recommendation system for articles. Chase feed engagement and you get a feed full of slop. The scanner is just them selling a fix for the thing the strategy guaranteed.
I don’t follow or engage with anyone here because of the number of times then drop a Note. I follow because of shared interest, and the value and quality of the writing. Quite the opposite, I am more likely to unfollow a writer when they flood my feed with short Notes and Restacks.
I am sick to death with hearing about “growth” from every corner. Making growth the goal is one of the most pervasive causes of enshittification. It is a fabricated metric which, in this case, benefits SubStack. In my opinion, writers are, perhaps unintentionally, complicit when they buy into the metric that the platforms dangle in front of them.
For the record, your posts (and notes) have always registered as 0% bullshit. That is why I follow you!
I left out the part about complicity. The platform creates the metric, but it's up to the writer to use it, and most do, since seeing the number rise feels like proof of success. But who actually benefits from this growth? The writer who chases it and the platform that records it are not the same, and the difference between them is where things get messy.
Thank you. Since you usually unfollow at the first sign of too many posts, your words mean a lot.
There is nothing bad in using “AI as a tool”, especially if one uses English as second (or third) language. I’ve always checked my grammar against “the net” and searched whether the sentence I wrote had been used in a similar way somewhere else in the English-speaking world. I have never used Grammarly, because after all I do not mind being identified as a non-native speaker (I am!), but I’m not against.
What I completely dislike is 100% AI-generated articles. I understand that people “research” the sources (and… well what is research these days? Maybe a prompt in some AI chat?), feed the sources to some AI, generate an article and (in some cases) review the final draft, but what you get in many cases is that kind of stereotypically-written article whose rithm and wording make me uncomfortable.
Not to mention the AI-generated articles in Italian (my native languages): in that case (especially if you are fluent in English) you can perceive a kind of non-existent neo-lingua that nobody taught you in school and has nothing to do with real Italian. Words are correct and used properly, syntax and verbs are correct, but the whole “thing” (how to call it?) sounds completely fake. I hate it.
So to me it is more about the form, not the content. Writing for people in my opinion should be done with the aim of engaging emotionally with people, not just for conveying information. Nowadays if I need mere information I turn to Google or BBC or… an AI prompt. I am convinced that one can add a human touch to everything, even to an article about “dictionaries in Python”. Maybe it is a slower way of writing, maybe you will be getting less revenues at the beginning, but it will pay off in the medium to long run.
Just my opinion. I appreciate this discussion. It is important to use new technology while not forgetting to be sentient humans.
The Italian example is the best thing in this thread. Every word correct and the whole thing still fake. The machine tries to pass as a native and misses the stuff nobody teaches you in school.
Funny thing is, idioms are where I get lost too. I never know if one fits or where it goes. Same spot the machine trips on, for the same reason: neither of us grew up inside the language.
Mentioned you and this post in mine
https://ombreolivier.substack.com/p/pangram-and-substack-why-the-whining?r=7yrqz&utm_campaign=post&utm_medium=web
I really don't care if anyone "uses AI," nor how they "use AI." All I care about is whether the result, however generated, is bullshit or not. No AI scanner can detect bullshit, because all AI can produce is bullshit, since it has no understanding of meaning of the words it produces. Only a human can detect bullshit -- either a human using AI as a writing assistant, or a the reader.
SubStack's colossal bunder is deciding that what was needed was to integrate an AI detection AI (how incredibly absurd this sounds) into their platform. If someone wants to use one of these tools to scan the post of an author, then there is nothing stopping them from copying it and pasting it into the scanner of choice. Have at it -- if that is what you want to do.
What no one needs is to have the platform "score" posts using a built-in scanner.
Enshittifcation, it seems, has arrived.
The enshittification started earlier than the scanner. It was the day they decided growth meant Notes volume instead of a real recommendation system for articles. Chase feed engagement and you get a feed full of slop. The scanner is just them selling a fix for the thing the strategy guaranteed.
Growth for whom, writers, or SubStack?
I don’t follow or engage with anyone here because of the number of times then drop a Note. I follow because of shared interest, and the value and quality of the writing. Quite the opposite, I am more likely to unfollow a writer when they flood my feed with short Notes and Restacks.
I am sick to death with hearing about “growth” from every corner. Making growth the goal is one of the most pervasive causes of enshittification. It is a fabricated metric which, in this case, benefits SubStack. In my opinion, writers are, perhaps unintentionally, complicit when they buy into the metric that the platforms dangle in front of them.
For the record, your posts (and notes) have always registered as 0% bullshit. That is why I follow you!
I left out the part about complicity. The platform creates the metric, but it's up to the writer to use it, and most do, since seeing the number rise feels like proof of success. But who actually benefits from this growth? The writer who chases it and the platform that records it are not the same, and the difference between them is where things get messy.
Thank you. Since you usually unfollow at the first sign of too many posts, your words mean a lot.
Enshittification on Substack has been underway for quite some time.