The system of AI testing repetition rate violates human rights

Philosophy Forums Society, Government, and Economics
1

I noticed that this website is very averse to Ai, so I wrote a paper about Ai, hoping to inspire this website.
The core idea is as follows:

The system of AI testing repetition rate violates human rights.

The logic is:

You have the right to take a car, and you also have the right to walk. If someone does not allow you to ride or walk, they are violating your human rights.

The same principle applies:

You have the right to write papers by hand, and you also have the right to write papers using AI. If someone does not allow you to write papers using AI, they are violating your human rights.

Abstract of the paper:

Contemporary academic evaluation faces a profound crisis of legitimacy. Textual

screening systems, designed primarily to detect AI-generated content, have reduced

academic evaluation to little more than an archaeological inquest into the instruments of

writing. This paper introduces “tool-provenance bias” as its core critical concept. Drawing on

the diagnostic framework of the Three Principles of Eudaimonia, it systematically

demonstrates the threefold harm inflicted by current screening regimes: the conflation of tools

with ends, reliance on unreliable detection technologies, and the subordination of substantive

scholarly contribution to formal compliance. Through historically nuanced re-examinations of

Mendel, Wegener, and Galois, this paper argues that the “textual provenance” paradigm

impedes the progress of knowledge. By engaging with Kuhn’s challenge of

incommensurability, it establishes the epistemological foundation for an "argumentative

validation" paradigm. Through comparative analysis of utilitarianism, Kantian deontology,

Rawlsian justice, and virtue ethics, it demonstrates the distinctive advantages of the Three

Principles of Eudaimonia. The central thesis is this: academic evaluation should shift its focus

from “who wrote this” to “whether the argument holds”—from textual provenance to

argumentative validation.

Keywords: academic evaluation; paradigm shift; Three Principles of Eudaimonia;

argumentative validation; textual provenance; knowledge progress

The detailed paper can be downloaded here:

https://philpapers.org/rec/TPSGGL

May I ask if this paper can be published here?

1 Like
What are cults and how can you know you are in one?
2

But, because you do not have ‘a right’ to abuse things, you do not have ‘a right’ to take a car.

First you will have to present the sound and valid argument/s that you human beings have ‘rights’, and from who or where ‘those rights’ have come from, exactly, before you can ‘harp on’ about ‘human rights’, and what they are or are not exactly.

So, have you got any sound and valid arguments here?

3

I dont think anyone would object to publishing it here.
After taking a look at it though its extremely technical for lil old me.

Of course the statement in and of itself is true, but also picks a
 rather curious angle.
I feel like there are several things we jump over before we get to the idea of someone trying to ban you from using a tool, if given the nature of the tool poses no danger to yourself of or the public.

Putting nutters with moodswings, emotions and opinions aside, what would be the opposition’s argument here? Of taking the ability of someone to use a specific tool.
That would require justification.

So
 in laymans terms i kinda feel like you took a defensive position before there was even an attack.
Cause i dont see a valid foundation for trying to take your right to use an AI.

1 Like
4

I agree with the thesis. I had an older brother who had a Phd in the area I was getting my Master’s. He wrote my papers. Obviously he’s not just a tool, but he was a tool in this instance for me. His work, under my name, should be judged on its merits.

I’m surprised we are still dealing with this issue around AI. There have always been incredible tools available.

I also agree that AI detection programs can be fallible and it allows suspicion of students and academics. Before the era of AI and tools to detect it, suspicion was not aimed at student’s work and any tools used to analyze for possible plagiarism or other cheating were absolutely perfect, the professors being infallible and fully trusting.

Some people will foolishly say that passing academic thresholds can lead to roles and jobs where your own competence and understanding of the arguments and fluency in the field make you competent for those roles. But that is discriminating against people who love learning for its own sake. A paper is a paper, not part of some process of context.

Our cousin is in a med program and he is allowed to answer doctors’ questions using a very powerful AI he has in his mobile. It monitors the conversation and when Gerry is addressed, gives him right answers at a rate higher than his peers. Let’s repeat this important point: he is correct more often than his peers.

We want doctors who can get correct answers.

5

Sarcasm aside, I think paper writing is some weird academic parasitical hurdle that’s not necessary for most jobs, except if you are going into academia. And that is basically circular reasoning. For most jobs - including those that generally require people with degrees - you don’t need to be able to write good papers to do the job well. You really don’t.

My suggestion would be that one can take two different tracks or programs through the same courses and degrees. You can take courses with the paper writing because you might expect that you will do research and publish papers, so you want that as part of your future skills. And then others can take non-paper writing version of the same courses. And maybe they’re even in the same classroom with the same professor with different testing protocols. And then if you really need to learn how to write papers later, then there’ll be certification classes you can take. I don’t know how academia’s strange culture got so much power that everyone has to go through this boring useless to them hoop.

1 Like
6

You don’t have to write good papers, not even for research.

Some people are just good at making papers look good, not the content.

1 Like
7

If you had actual awareness of the cult that you are in, then you would also have, at least some, some awareness around how that cult/ure got so much power, that you and everyone else went through the hoop, and have “chosen”.to stay in that loop.

8

Anew

  1. performs ‘mindreading’ here in an implicit ad hom.

  2. continues to bear his grudge

  3. hints at his special secret awareness, but does not share it

  4. assumes a bunch of stuff about another poster rather than adding substance to the thread.

  5. does not respond to any points made by the person he is responding to

I’ll ignore any further off-topic posts by him in this thread also.

9

LOL How many times can someone respond to posts claiming to ignore any further off-topic posts?

What we have here is further proof of one who just does not recognise the cult that they are within.

10

There is no law violation for using AI to write a paper or an essay. If such rule exists in a website, it is in terms of business policy. Every private business can have special rules. Think for instance luxury restaurants that have a dress code. There is no law forcing you to wear suit, yet these restaurants will not allow you to enter without proper attire.

AI is powerful tool, but there are several issues with repeated use or it. Many people do not appreciate talking to machines. Also, if I use AI to write an entire paper with zero personal contribution, then this paper is useless. Pupils that use it frequently do not have the chance to develop critical thinking skills, since all the thinking is done by chatgpt and the likes.

We are in a transitional period where AI threatens job positions. It is understandable that people want to put boundaries in the use of AI.

11

I never went through that hoop. But as usual anew:

  1. assumes things

  2. focuses on people not ideas

  3. has nothing to say about the topic

  4. tries to signal his ‘superiority’

12

LOL, LOL, LOL

So, once again, ‘this one’ believes that it does not do what it claims everyone else does.

Why do you keep seeing ‘my superiority’?

13

I agree with your viewpoint
Engaging in scientific research is like playing football. Some people are defenders, some are forwards, and some are goalkeepers. Only through division of labor and cooperation can this team perform well.
The same goes for conducting scientific research. Some people specialize in innovation, while others specialize in writing papers. Only by dividing labor and collaborating to build a happy environment can this team be efficient.
For detailed information, please refer to my paper "

Proposal on the Reform of Doctoral Education

"

14

Ah, good catch. I wrote incorrectly. It should have been more specific 
everyone who goes through traditional academia. Some go through non-traditional where they don’t use essays. Many do not go through academia at all.

I don’t see your superiority or ‘your superiority’. I see you signaling that you are superior. I was a wee lad when I realized advertising was not hinged to the truth.

15

We’re both critical of essay writing as a central hurdle in academia. We disagree about your argument in the paper. If essay writing is one of the main academia criteria for passing a course, then students should not be allowed to use AI, just as one was not allowed to get other people to write your essays pre- AI. You need know nothing at all when getting an AI to write the essay. I think it is a poor evaluation device, but it is an evaluation device. If you remove evaluation devices people will get through who need know nothing about their subject. You could make the same argument as you did for using AI with papers for using them on tests and in group discussions. Essentially you would be allow students to get degrees with near zero knowledge of their majors, Master’s and Phd topics.

My suggestion to academia is they remove papers from the evaluation process and find other better ways to evaluate student knowledge and skills. But if it is part or the main part of the evaluation process, letting students use AIs is both utterly meaningless and removing an requirement that at least shows something about what they know, however poor a tool it is.

16

Before the invention of excavators, laborers had to dig earth with hoes by hand. With the advent of mechanization, evaluation criteria evolved. Assessments no longer tested the ability to dig manually; instead, they focused on machine operation and engineering planning. As people’s skills advanced, evaluation items kept pace with the times.

The same logic applies to AI as a production tool. In the past, academic competence could only be tested through fully handwritten manuscripts. Today, with technological progress, academic ability is no longer confined to text drafting and transcription. A researcher’s core competence lies in posing questions, constructing argument frameworks, checking logical consistency, and revising viewpoints, while AI merely polishes wording and drafts manuscripts.

Once the focus of evaluation shifts from manual writing to ideological reasoning and logical deliberation, the use of AI cannot be equated with cheating. It is no more a violation than an engineer making use of construction machinery.

揑è‡Ș我的 iPhone

17

No. This is not correct. You can go to an AI and prompt it try to find a topic that hasn’t been written about much. You give it the assignment, whatever criteria you have been given by the teacher/professor. You ask write the essay with the topic it has chosen - which you can run by some other AIs to see if they agree it is a less common but interesting topic. Ask it to use the proper formatting, and give it a length and any other criteria it needs to meet. Then you take it and you put the results into another AI, and you have it check through, check what’s written in there and fix any mistakes/tighten up weaker arguments, etc. Then take it back to the first AI see if it approves the changes - you can use a third AI if there is disagreement. And then you ask it or one of the AIs to modify so that it does not seem like it is the product of an AI. Then you run the changes against by another AI. Tweaking, testing, etc. You can also ask it to include a few weaker arguments, missed commas or the like so it doesn’t seem too perfect. Aim for a B+ or a 3.6 or whatever. You never need to read it. You don’t need to understand the arguments. You don’t need to understand a word of vocabulary in the field of research that it’s in. None of that. You don’t even need to know the topic, though you might accidently read that while copy pasting the topic and essay around between AIs. If you are a digger, you need to know how to dig with a shovel. If you dig with an escavator you need to know how to use one. When escavator drivers are replaced by AI drivers, there’s no job for a human. But if the degree leads to work that one knows nothing about there is a serious problem.

18

The only way I see to avoid AI use when you test students skills is the old fashioned oral examination. Students physical presence being obligatory and they should provide answers in questions in real time.

With the current number of students per class, this is not something practical (almost impossible in many cases). Yet, how else can we verify students’ abilities?

19

I think oral tests is better than papers, especially if you don’t have to be in some kind of formal debating, perfectly organized speech. You have a discussion with the professor or TAs or a group as if you were colleagues discussing topics relevant to the subject and any special area you focused on. Yes, oral tests would take time, but if you eliminate lectures, which I think should be rare things by visiting experts, not regular pedagogy - then professors have a lot of time suddenly. It would make for dense end of term work periods and lighter work before then.

Another way to do it is via Problem based learning groups. You split the students into groups that work together. They evaluate themselves and others and the teacher can attend meeting, which can happen during what would have been lecture time. Lectures are a terrible pedagogical tool. The teachers, you and your peers can see you in real time engaging with ideas, experiments, reporting on research you’ve done, etc.

Tests can work. Short essay answers, for example. Not essays with all the structural, formatting, footnotes and so on stuff. But where you write a paragraph answering questions. Even traditional multiple choice tests can work for entire subjects like anatomy, but also parts of the knowledge base for other subjects.

20

Could you be wrong? Or, is that not a possibility, in ‘your world’?

Okay, but if you could be wrong above, which I know what the truth is, then what you realised here when you were what you call a “wee lad” is moot anyway, right?

Also, and by the way, sometimes advertising is actually hinged to the truth, but obviously does not have to be.