Hey, I'm currently writing an assignment that I have to hand in on Friday. Since everyone was talking about teachers checking whether the assignment was written by Chatgpt, I decided not to bother. Now I've tested what Chatgpt says about my previous work, and it simply said that the ENTIRE TEXT was written by it. Hello? I'm risking a grade of 6 for this crap.
(2 votes)
Loading...Hello dear ones, I created a Discord server with my best friend, and we both noticed that when new people join, you can't see previous messages. (Actually, you can see everything previously written in the channels, as well as rules, reaction roles, etc. I've also pinned the rules and roles.) I'm now wondering why this…
The public type Example must be defined in its own fileJava(16777541) ???
Hello, I have a photo on my desktop. When I right-click and select "Delete," it doesn't respond. How do I delete it?
Hi, I need thumbnails for my videos and I came across Chat GPT. I think the thumbnails are really cool too, but there are always spelling mistakes in the font. Do you know what I could do about it? Here is an example: Correct text: Build Iron Farm
Hello, I'm working on a project in JavaScript. I'm working on a calculator. I want an image of Pi to correspond to the value of Pi so that I can calculate with it normally. Here is my code. function calloutpi() { var container = document.getElementById("resultAra"); var img = document.createElement("img"); img.src = " https://th.bing.com/th/id/OIP.K_JHbZJ4pIeZOGHtMR6U0AHaES?pid=ImgDet&rs=1 "; img.value=Math.PI…
This is simply because ChatGPT has no idea what you could recognize a text created by ChatGPT.
The basic problem is once again the lack of understanding what ChatGPT is. It is a language model, which means that it is very good to build sentences that match a given question or request. The language model itself has no idea of logic or how the world works. While ChatGPT is partially underlined with facts so that there can be meaningful answers at least to basic things, the AI regularly gives false answers. Because she’s just good at writing sentences that sound good. She’s not omniscient.
OpenAI, the developers of ChatGPT themselves have tried to develop an additional tool that should be able to recognize texts created by ChatGPT as such – and failed.
ChatGPT has now learned writing from people, so it also writes like a person. What are you supposed to know if it was written by an AI or by a human being? There are now only words that have come together, like those I write here. There are no hidden data or patterns. Only ChatGPT’s writing style could be recognized, but on the one hand there are also people who write so, and you can also ask ChatGPT to use a different type of writing or to insert typing errors. Therefore, neither human nor machine can say exactly whether a text was generated by a human or by an AI.
ChatGPT lies. If the answer doesn’t know, there’s something in it. Ask him a question, and then say the answer is wrong, even if she’s right. ChatGPT will give you another answer. The main task seems to be that the questioner receives an answer that satisfies him. ChatGPT then works even if the questioner gets an incorrect answer. He mustn’t know.
If your teacher has the slightest idea, he won’t just ask ChatGPT directly if the text is from the AI. Because everyone who played something with it knows that it can say something.
More important when dealing with ChatGPT is that it is just inventing things, and therefore simply should not be used. There’s wrong information.
If one tries to clean the window with a hammer, it can lead to catastrophic results. However, the solution for this is not to use hammers in principle, but simply to use them for what they are suitable for.
ChatGPT can do many things pretty well. But research isn’t part of it. But when it comes to simply writing texts, it is an ingenious tool. For example, you can specify keywords and a style, and it writes a text. If the text then contains all the desired information and makes sense, you know yourself. And even if ChatGPT itself cannot research well, it can be a useful search tool. There may well be a rough overview of a topic, give points or suggest further sources of knowledge. This allows you to continue researching yourself and discover facts without relying directly on ChatGPT.
It was sloppy. I meant ChatGPT shouldn’t be used by just letting you do the homework without controlling anything. It is a tool, you have to know what is possible and use it accordingly.
Thanks for the correction.
It is not possible to check with certainty whether the text originates from the AI. Yes, there are applications that are unreliable. He shouldn’t give you six.
If I had considered as a teacher, I would ask the pupil to answer questions. You can see if he’s been busy or not.
ChatGPT cannot be asked. The tests are different.
It is more similar to a plagiarism examiner and he is looking for clear text features.
You should be aware that ChatGPT is not a human being and therefore ask questions not truthfully answered, lies or simply has no idea what you want from him. In the latter case, it just gives you a wrong answer.
The company of ChatGPT (OpenAI) has brought out a tool to see if a text was written by ChatGPT.
But they have set it up again because it is in approx. 75% of the cases were next and many false positives delivered.
Right. But this is not ChatGPT itself and apparently the questioner asked ChatGPT.
Oh, okay, my mistake. Thanks for the info
I don’t think you can do anything, try to explain it to your teacher…
LG, Regagi1038!
You assume ChatGPT can recognize the difference.
That’s your fault.
And you assume that teachers use ChatGPT for review.
That would also be a mistake.
Do you have the language course? I’d be interested.