Install Steam
login
|
language
简体中文 (Simplified Chinese)
繁體中文 (Traditional Chinese)
日本語 (Japanese)
한국어 (Korean)
ไทย (Thai)
Български (Bulgarian)
Čeština (Czech)
Dansk (Danish)
Deutsch (German)
Español - España (Spanish - Spain)
Español - Latinoamérica (Spanish - Latin America)
Ελληνικά (Greek)
Français (French)
Italiano (Italian)
Bahasa Indonesia (Indonesian)
Magyar (Hungarian)
Nederlands (Dutch)
Norsk (Norwegian)
Polski (Polish)
Português (Portuguese - Portugal)
Português - Brasil (Portuguese - Brazil)
Română (Romanian)
Русский (Russian)
Suomi (Finnish)
Svenska (Swedish)
Türkçe (Turkish)
Tiếng Việt (Vietnamese)
Українська (Ukrainian)
Report a translation problem
Discussions Rules and Guidelines
Report this post
and i mean, what do you even doing at this point? the professor is not going to believe you, regardless of if you tell the truth.
i think your best bet is to write notes as you research your assignments, have clear references (books, websites) with page numbers/passages/links all detailed where you obtained your information. give yourself a long receipt, so you have plenty of proofs that everything was done by your hand and mind alone.
I find it similar to a polygraph test. Good for putting pressure on the suspect, but terrible at providing actual evidence of malfeasance.
In your case, it's entirely possible that the algorithm treated quotes and citations like any other part of your essay. Context isn't something AIs naturally intuit. So on the off chance that they ask any questions, as long as you're honest and open you should be fine. And if you do use AI to supplement your work, it should always remain an auxillary tool.
Academic writing is problematic in particular because most language models are trained heavily on that kind of content. Write in an academic way, or pepper your language with rhetoric, and these so called AI detectors will see GPT-isms everywhere and claim you're a bot.
Then it will check the perplexity, burstiness, low levels of randomness in word choice, sentence length and so on.
Due to even my self-taught photographical memory recall, I can be sometimes detected as AI. Yet that's just due to reading something online and recalling it.
So yes, citations and quotes commonly are false positives and assumed to be AI written due to drawing it from online "matching" content. It's quite stupid that way.
Almost all AI Detectors aren't that accurate and not reliable at all, with a high level of both false positives and false negatives, just depending on how you write and structure your sentences.
Ai Detection: 0% of text is likely AI
(pro-tip, writing your academic essays like Yoda, you should start; nvm, probably won't help)
then the teacher takes that number and uses it to try and justify their pre-existing biases for or against the student, and makes their determination about how to respond.
In short :
It's a black-box akin to a magic eight ball, that leads to students either being eliminated at random, or helps teachers (that are intentionally trying to ruin certain students) justify their prejudices.
Make sure to show this topic to your Dean of Students. :)
Unless you're not in higher ed yet, then uhhhhh.... well, I think if you're somehow able to use an A.I. to generate a paper in Elementary school, then something is probably very wrong with how the class coursework is being managed. And in junior high or high school... idk, start with the vice principle if you get called a cheater over this and then get your parents' help to escalate it upward from there.
Because the cited materials are going to be in the training data for the A.I.
So, now you have an ugly situation where you need citations for the assignment, but you need to actively do the bare-minimum amount of citations in order to protect yourself against some black-box "detection tool".
I had some skepticism with plagiarism detection tools being black-box too, but at least those were using trusted databases that could be audited by trusted professionals.
The A.I. detection tools aren't like that, though.
This is literally just one A.I. saying that it thinks the thing submitted to it was written by an A.I. with the same training data as the detection A.I.
The consequences go WELL beyond false positives, because that too will lead to a cascade of further consequences, as I'm about to suggest :
Higher ed hasn't believed in due process for a while, but they're just now pushing it to a point where most people should seriously consider breaking all the rules and using underhanded tactics such as blackmail (as a last resort) to proceed. (If they are able to.)
(This shouldn't be your first-strategy by the way.)
(It shouldn't even be your second strategy
unless you've tried it before and it has worked well, I guess.)
If they're not going to bring an evidence-based case against you,
and if they already consider you a cheater before you've even begun,
then why try to be anything but?
( I hope you're thinking deeply about this question, Mr. Dean of Students.)
( Welcome to the world that your professors are creating ! )
The fact of the matter is,
the question of "how can we prove that our students used A.I.?" is a question that doesn't make sense,
and only comes from teachers who would rather police instead of teach.
As a student, you cannot fire such teachers that ask such hostile and irrelevant questions;
they stand in your way and threaten everything you've been working towards when you come across them.
The brave new world we're entering requires that you be looking for ways to ruin people who you were supposed to respect. ...before they can ruin you.
We're not quite to the point in this next video yet
(mainly because LLMs are not true A.I.)
but it doesn't matter much, since, the way that people are running large organizations -
(such as colleges, universities, police forces, and corporations) while using the technology without thinking about what the purpose of their jobs are -
is leading to alarming situations that are CLOSE ENOUGH :
"victim or perpetrator, you stand in its way"
Whether it be an A.I., or an administration, or merely a supervisor,
And this is primarily brought on simply by authority figures not implementing & upholding due process rights within the punitive-measures aspect of their organizations.
However, in the case of Academia, this can generally all be avoided by teachers simply staying in their lane, and TEACHING instead of policing.
Classrooms aren't the place for handcuffs.
And the answer to that would be,
"I don't care because I'm not trying to be them. I'm trying to be me,
and I've had to face HUMAN adversaries that they have not!"