Secret Sabotage
The other day my wife got a text from our sister in law about an incident that happened at our nieces school. She actually sent it to my wife and my sister in law since they both work in the education system, but given the fact that I do as well, I got pulled into it. I'd like to say this is the first time I have heard of something like this happening, but sadly, it is not. On top of that, I think incidents like this are going to happen more and more in the future.
You see, our sweet, innocent, loveable niece was accused of using AI to write one of her reports for school. Either that or she was accused of having her parents write it for her. It was never quite clear what the teachers insinuation was. Which as I have posted about before, brings up one of the many issues with AI.
The Seven Point Structure
If you do a quick Google search, it isn't too hard to find many pages referencing the Seven Point Story Structure. This is a method of writing that has been taught for ages at the grade school level. Sometimes they simply it down to four or five points, but the basic premise is the same. There is a opening paragraph that is considered the starting point or hook of the paper. Then you have several points in the middle before finishing with the conclusion.
This is a basic formula for writing that can be used for stories, but also adapted with a few changes for professional writing and term papers/essays. That's likely what most of you are familiar with unless you took a creative writing class in school.
The problem is, since it is widely accepted and abundantly accessible on Google, AI knows about the seven point structure as well. In fact, most AI created content utilizes this process or some variation of it. It's one of the ways you can usually spot AI created content, until of course you come across someone who knows how to use the structure. Then you need to dig a little deeper.
It turns out a AI pass concluded that the piece written by our niece was over 90% not AI generated. That should have been the end of it right there, but it ballooned into a big thing due to some poor decisions by the school that I will not get into here. The fact remains, AI is disrupting even the most mundane things of everyday life, to the point that even when you do things "right", you can still be accused of cutting corners.
Meanwhile I guarantee the adults use AI to do half their work.
I saw another story on Reddit last night which I can't seem to locate again now, but it was talking about people using Canva for their resumes. More specifically Canva templates for their resumes. Now, I'm a huge fan of Canva. I think it is a pretty amazing tool, and I have been known to dabble on the AI side of it a time or two given my absolute lack of creativity.
Anyway, this post had a recruiter talking about how they had a stack of resumes (it's called something else in Europe, CV I think), through an AI program to separate them or something. Anyway, it turns out that CV's generated via a Canva template have metadata attached to them that doesn't necessarily line up with the actual text.
I can't verify this of course, but in this case the writer was talking about how the CV from this nursing candidate in Michigan was being sorted based on the metadata into a stack of resumes for construction workers in Idaho or something like that.
If you understand metadata, it's not surprising to hear that something like this might be happening. Let's say that lady posted her resume to several job websites like "Indeed", and let's also assume that those sites use AI to separate and filter submissions because who apparently doesn't these days...
Think of all the job opportunities she may have missed out on due to her resume being classified wrong based on the metadata from the template versus the actual content of her CV.
Scary right?
It's almost like we are secretly sabotaging ourselves by using these tools. Even more so given the fact that most people don't understand how they work, but they are trusting them with basically every aspect of their lives. Even worse, we have the case of our niece where even when you are doing everything right, you still get the short end of the stick.
Our niece did end up getting an "apology" from the school, but I read it and it was one of those apologies that wasn't really an apology. Not I'm sorry I screwed up, but I'm sorry you felt bad about this...
Typical!
My Sports Account - @bozz.sports
Come hang out in The Pub
https://bsky.app/profile/did:plc:u37t33dyaufvrqujwmvm54ln/post/3mhfut5gpmk2m
https://bsky.app/profile/did:plc:u37t33dyaufvrqujwmvm54ln/post/3mhfut5gpmk2m
The rewards earned on this comment will go to the author of the blog post.
The apology was probably AI written by administration, hahaha. As an old guy, it's been interesting to watch as we went from spell check, to grammar check, to now AI written, edited, and now evaluated documents.....crazy!
I was thinking that too. It's only going to get worse, but hopefully I can put it off as long as possible.
We do like the latin word indeed. Curriculum vitea, I believe. Too lazy to look it up or asked AI.
I also use AI to structure some of my work. Or to find flaws in designs. I do like AIbut not the way many or using it. It should be a support tool and not be considered a co-worker.
Co worker or replacement for yourself as a worker I think. I think a lot of people don't understand the message you are sending when you obviously use AI for certain things.
If you would let AI do all the work for you, you probably would deliver sub par quality. If you use AI as a tool, your work probably would be above par.
But that’s my opinion.
I can appreciate that.
Yes, that's the word.
Phew those 3 years of Latin I did follow paid off finally 😁
3 years spent on learning a dead language you only use if you're a doctor or lawyer? 😆 Jesus! Why is Denmark torturing kids like this.
We only had one, but needed some terms for law courses at uni.
Not Denmark, Belgium.
Learning Latin was an own choice 🤦♂️
Ooops, I meant Belgium 😬 Sorry.
This is going to be an issue and more people are going to get hurt. It sucks when you're accused of something you haven't done, no one believes you and there's not much you can do to prove it.
Regarding job applications, I've seen a report in TV, about job interviews. There was a guy who applied for a certain position and as a next step, he was invited to have an interview with an AI bot. He said "no thanks, if you can't even bother to interview me in person, I'm not interested".
I think we're going to have to find a way to handle things the right way and find a balance with AI. I hope your nice can get out of this mess and also get over it as it can mark you, especially at a young age.
I think she will be okay, it was pretty clear based on some other info the teacher was just having a bad day and for some reason decided to take it out on her, but yeah, it still sucks. I think a lot of employers are going to start relying on AI and then being sorry after the fact about the quality of workers they are getting because of it.
Meanwhile we have changed our official strategy from SAS first, COTS second, Development third to AI Development only. We are at a three day from customer request for an app to the application being deployed to Production...
I don't know what it is like in schools now, but my kids are through that already and didn't have these tools. I didn't have to help them much with their work anyway. It must be tough for the teachers, but then the kids may be wondering what their future is when machines can do so much and will be evolving fast. It's scary.
It definitely is. I am kind of happy I am close to being down with my career because there is going to be a lot of policy stuff that involves tech but is not really the kind of tech I want to do.
AI is going to bombard us with endless arguments about originality, but it’s all about using it wisely—and yes, it’s already essential in many fields.
We’re going to have to live with that.
Great photos; thanks for sharing the topic and the images, @bozz
That’s harsh. Information are so wide spread, its going to touch some part of the internet.
Your last line sums it all.
It's all a little crazy as we start to understand the new tech. I do feel sorry for teachers who are really under the pump to educate, and being told to be vigilant for dishonesty - which has always been plagiarism, but now takes the form of AI - they're bound to stuff up sometimes. Then they're probably poorly educated themselves on how to handle it. How to teach kids discerning use of AI, for example? When it is it right to use it, and when might it be risking your ability to learn and to prove you can learn with old fashioned and rigid assessments?
Then we have AI producing content which mimics the content itself is creating, and to sound 'perfect' we have to follow this structure too - even though the structure you speak of (we don't call it the seven point story structure - but it's essentially that) can be played around with a little. So we're all churning out sameness along some made up perfection guidelines, which AI is copying, and we're copying AI, who is copying itself back. Sigh. Are we on drugs?
I was reading about a guy whose article was booted from a magazine as they said it was AI generated - he was a respected journalist and hadn't used AI at all. Yep, we've dug this hole.
Don't be too hard on the teachers using AI - they're so under the pump that if there's a shortcut, they'll use it. I don't plan lessons anymore but I did have a term's cover a few months back and honestly, getting an AI to write sample paragraphs to my guidelines, and giving them a quick edit for my purposes, saved me so much time it wasn't funny. Even just a quick few ideas for lesson plans. Honestly, if I'd been given more planning time, and wasn't teaching five classes, I wouldn't have needed it. We seem to be pushed into this hyper productivity that demands we do the work of three people. Rather than rioting, we're looking for short cuts to save our sanity.
Saying that, the school obviously needs a better approach. Certainly moderating it with more than one staff member, having an AI consultant/expert on it, and maybe a different way to assess might help? Same with the CV's - I mean, they just have to look snazzy to get in the door so you get the interview, right? That's where the real assessment begins - not on paper. And if the employer is dumb enough to assess CV's on the basis of font, well...
Your reply is upvoted by @topcomment; a manual curation service that rewards meaningful and engaging comments.
More Info - Support us! - Reports - Discord Channel