221–230 of 288 posts
Earlier quoted context omitted.
> And if student can not recognize their teacher is bullshiting them If it's a learner's first exposure to a topic what indicators can they use to identify that ChatGPT is lying to them?
The same indicators they would use with teacher? Studyig is about reading books, comparing sources. Not copying notes. At university if you discover teacher is lying, you may get prosecuted, expelled, and loose $100k on student loans. Edit: response to your other comment, I am being rationed by HN As I said, ChatGPT is good at answering and explaining questions. It can replace teacher as "explainer", not as some sort…
-- For each of the students.
-- For someone in the Computer Science department.
-- A different prompt tool (like Bard even)
I hope this terrible form of evaluation & judgement doesn't affect our legal system.
Earlier quoted context omitted.
Remember when people didn't trust Wikipedia? They actually taught us about references and primary sources. Then we all kind of forgot, or got lazy. Same thing with this, even if the bots list sources, we won't read them, we barely read what the bot says to begin with.
One of the possible outcomes of the current trends is that we may rediscover the need for a strongly curated knowledge base as a common reference of conversational truth. (E.g., a classic encyclopedia, advisably more than a single one.)
Honestly, I would love something like this. Sometimes I want to buy some sneakers, but all that comes up is Amazon and a bunch of trashy spam when I try to find something. Searching for things has become nearly impossible anymore.
Earlier quoted context omitted.
I don’t know much about him but I thought he was a researcher? There is an “Academic publications” section on https://en.m.wikipedia.org/wiki/Eliezer_Yudkowsky and it says “He is a co-founder and research fellow at the Machine Intelligence Research Institute” LLMs are not self aware, I agree, but there is clearly some reasoning going on and you could argue that its model of human behaviour can somewhat simulate self…
> its model of human behaviour What model of human behaviour? I asked ChatGPT if it has a model of human behaviour, and it said "ChatGPT does not have an explicit model of human behavior".
It just hallucinated an answer
If it didn’t have a model of human behaviour, then it wouldn’t work, because the whole idea is it’s simulating what someone acting as a helpful assistant would do.
Earlier quoted context omitted.
Seriously, though. Q to ChatGPT: ‘Was this text generated by ChatGPT? "The ability to 'cheat' on written assignments has always been a possibility. With the rise of the Internet, it became easier to source it from more places - with a counter balancing rise of plagiarism detectors. Now with Google, Grammarly and ChatGPT, it is trivial to copy/paste a pipeline with enough entropy to defeat many if not most detectors.…
I just had ChatGPT write me a poem, and then in a new session, pasted the poem back in and asked it if it wrote it. The response? "No, I did not write that poem. I am an AI language model, and I generate responses based on patterns and information in the data I have been trained on. I don't have the capability to write original poems or claim authorship of any specific text."
Clearly the poem was written by a different instance of ChatGPT! It must be very difficult to define a sense of self when you are a predictive text generator implemented as a distributed application running across multiple datacenters.
(of course there is no reason to expect a fancy autocomplete to have a sense of self in the first place...)
Earlier quoted context omitted.
> its model of human behaviour What model of human behaviour? I asked ChatGPT if it has a model of human behaviour, and it said "ChatGPT does not have an explicit model of human behavior".
As I just said ChatGPT is not self aware, so it can’t answer questions about it’s own workings like that. (this is somewhat true of humans too though, there is a lot going on in our subconscious, but we are aware of our own conscious thoughts to some extent at least) It just hallucinated an answer If it didn’t have a model of human behaviour, then it wouldn’t work, because the whole idea is it’s simulating what someo…
Is every answer thus not a hallucination by this logic? If it's trained on external information, why would this information not include observations and information made about it? A human doesn't need to be able to even read to present a book to someone else with contextually relevant information.
Is there a difference between cheating in school and hacking school? I think so. I almost never studied for math tests in University. They were all multiple choice and I found it easy to logic out the only possible answer to the question just by looking at the choices. Worst case if I was in a hurry I'd get a B on an exam. I wasn't looking up the answers and skipping any work. I was just arriving at the prescribed op…
How does an entirely multiple-choice mathematics exam at University level even work? What kind of maths? I don't see how any kind of proof or derivation question could work in this context.
Earlier quoted context omitted.
As I just said ChatGPT is not self aware, so it can’t answer questions about it’s own workings like that. (this is somewhat true of humans too though, there is a lot going on in our subconscious, but we are aware of our own conscious thoughts to some extent at least) It just hallucinated an answer If it didn’t have a model of human behaviour, then it wouldn’t work, because the whole idea is it’s simulating what someo…
Self-awareness isn't a requirement for the presentation of information, which LLMs continually prove. Is every answer thus not a hallucination by this logic? If it's trained on external information, why would this information not include observations and information made about it? A human doesn't need to be able to even read to present a book to someone else with contextually relevant information.
But this is true of humans in many cases too
It’s training data is from 2021, so it won’t contain anything about how ChatGPT works, maybe a bit about how LLMs in general work though
Earlier quoted context omitted.
The threads show that ChatGPT claims to have written anything if you ask it to. So it really doesn't matter what the student submits, ChatGPT will claim it?
Indeed there’s one screenshot showing ChatGPT claim it wrote the professor’s email. How embarrassing.