Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

CheatGPT

blog.humphd.org

Post reply on HN
Print view View original

371–380 of 544 posts

Re: CheatGPT

#371

Post by iameli »

hathawsh wrote: post #355
iameli wrote: post #284

Earlier quoted context omitted.

People seem so convinced of this and I just don't get it. I'm seeing this comment through my eyeballs, generating some pertinent text in my brain, and outputting it back out. But so many people seem convinced convinced this process is something radically, fundamentally, irreducibly different than what ChatGPT is doing internally, and I don't get why. Is it because I have a consciousness with an internal narrative and…

It's because you and I also have mental resources that are structured very differently from ChatGPT. Even if some part of our brain might resemble a LLM, a LLM is a very poor representation of other parts of the brain. If you take ChatGPT too far, it falls into repetition and demonstrates that it obviously has no comprehension of the material it's remixing. Maybe some combination of a LLM and other mental machinery w…

I dunno - I hear what you're saying about how it seems to lose track of things over long conversations. But part of the mental machinery that it lacks is the ability to learn from those conversations and insert new ideas into its base model; every new conversation starts fresh. It has no long-term storage; it can't "keep notes" for itself, and we don't give it the ability to alter its fundamental state based on new input after we're done feeding it training data.

I guess we're mostly agreeing, because those things are probably part of the "other mental machinery" we'd want to provide it, but I'd push back a little on "real comprehension". It kinda seems to me like the "comprehension engine" is working just fine, it's the structure we've built around that comprehension engine that's limiting it right now.

Re: CheatGPT

#372

Post by imperio59 »

I think what this means is that we are likely to see a next generation of software engineers that heavily lean on AI to "learn" programming and we will have to re-train them to some degree when they get into the workforce to be able to actually know the skillset they supposedly learned.

It also means that whiteboard problem solving for coding interviews is going to continue to be a great test to separate out students who actually know how to write a basic for loop and function and those who don't because Co-Pilot "always does it for them".

I know which kind of new grad I want working with me...

Re: CheatGPT

#373

Post by hombre_fatal »

haolez wrote: post #216
throw310822 wrote: post #209

I don't understand how most of the comments here seem to be along the lines of "these interview questions are useless now" or "we need to rethink education, it hasn't kept up". These all seem absurdly myopic to me. What we're seeing is the first instance, still very limited and imperfect, of AGI. This is not going to make some interview questions obsolete, or give students more tools to cheat with their homework. It…

I don't know if this is an AGI-like experiment, because LLM are trained on human knowledge. I'd expect that real AGI wouldn't need such a thing and would improve on its own. That's the moment where we become obsolete.

"you can invent human cognition from first principles via billion years of parallel evolution. or since it’s already been invented, applied, recorded at scale you can just observe its behavior to learn it"

Re: CheatGPT

#374

Post by barking_biscuit »

kurisufag wrote: post #222
throw310822 wrote: post #209

I don't understand how most of the comments here seem to be along the lines of "these interview questions are useless now" or "we need to rethink education, it hasn't kept up". These all seem absurdly myopic to me. What we're seeing is the first instance, still very limited and imperfect, of AGI. This is not going to make some interview questions obsolete, or give students more tools to cheat with their homework. It…

>What we're seeing is the first instance, still very limited and imperfect, of AGI absolutely not. language-model text generation is actually about as non-general as it gets -- they are fundamentally incapable of understanding anything at all, ever. they can't do math, work through basic logic problems, or produce any output that isn't just an assumed logical continuation of the input.

What about Toolformer?

Re: CheatGPT

#375

Post by andrepd »

adastra22 wrote: post #241
kurisufag wrote: post #222

Earlier quoted context omitted.

>What we're seeing is the first instance, still very limited and imperfect, of AGI absolutely not. language-model text generation is actually about as non-general as it gets -- they are fundamentally incapable of understanding anything at all, ever. they can't do math, work through basic logic problems, or produce any output that isn't just an assumed logical continuation of the input.

Have you actually used any of these products? GPT et al are perfectly capable of taking knowledge from any one domain and applying it towards the solution of any other problem domain, through various kinds of data abstraction, reasoning by analogy, and other techniques similar to what humans do. It makes plenty of goofups along the way, just like humans do. But if your requirement is that it performs absolutely perfe…

Nope, they are good at prete ding they do. Ask it something that can't be googled in 30s and you will see this.

I'm skeptical you can simply scale up this approach to full AGI.

Re: CheatGPT

#376

Post by taneq »

epicureanideal wrote: post #162
josh2600 wrote: post #123

So most interview questions like "make a binary tree" are dead. The best interview question that will never die: "What's the weirdest bug you debugged? What made it weird?" For posterity: https://www.gamedeveloper.com/programming/my-hardest-bug-eve...

> "What's the weirdest bug you debugged? What made it weird?" If you can remember, 5-10 years after you solved it.

If you don’t have at least a couple of weird enough debugging war stories to still remember years later, you probably don’t have very broad debugging experience.

(Not to say you aren’t a good developer, but you haven’t seen a lot of weird edge cases.)

Re: CheatGPT

#377

Post by boroboro4 »

JohnFen wrote: post #317
adastra22 wrote: post #241

Earlier quoted context omitted.

Have you actually used any of these products? GPT et al are perfectly capable of taking knowledge from any one domain and applying it towards the solution of any other problem domain, through various kinds of data abstraction, reasoning by analogy, and other techniques similar to what humans do. It makes plenty of goofups along the way, just like humans do. But if your requirement is that it performs absolutely perfe…

> through various kinds of data abstraction, reasoning by analogy, and other techniques similar to what humans do. No, that's exactly not how LLMs work. They are extremely good at predicting what sentences resemble the sentences in their training data and creating those. That's all. People are getting tripped up because they are seeing legitimate intelligence in the output from these systems -- but that intelligence…

I’ve just asked gpt3 to sum two large random numbers and it gave me correct sum of them. Then I’ve defined fibanachi like sequence (f1=1, f2=1, fn=fn_1 + fn_2 + 7) and it correctly gave me the value of 10th element. It’s not just statistical model to generate something resembling training set, it does understand training set, to similar extents as we understand world around us…

Re: CheatGPT

#378

Post by djleni »

JohnFen wrote: post #333
zeckalpha wrote: post #329

Earlier quoted context omitted.

Can you point to where the thinking happens in a human?

Nope. But I don't have to do that to understand that LLMs do not assess truth or accuracy of anything.

I don’t think it’s a stretch to say humans aren’t great at assessing the truth or accuracy of anything either.

Re: CheatGPT

#379

Post by dinkumthinkum »

corbulo wrote: post #79
dcow wrote: post #71

It seems ChatGPT should be treated the same way institutions treat plagiarism. It's fine to use LLMs for inspiration but the work you submit must be meaningfully your own. Of course it's not perfect, it never will be. But ultimately it's the student that suffers when they plagiarize. Professors ought have this conversation with students at the same times they're discussing the value of learning & education. I suspect…

It will be a hard sell since it or similar tech has already been long embraced commercially. A bit like forcing students to use an abacus while calculators are around. The student should still be responsible for what they submit and all of its errors of course. But what should really change is our methods of assessing students. The University model for undergrads has already been long broken in need of fixing. I don'…

Learning is different from commercial work. Ironically, it takes learning yo realize that.

Re: CheatGPT

#380

Post by dinkumthinkum »

ttul wrote: post #90

The best counter to ChatGPT cheating that I’ve heard of is to get students to orally defend their assignments in class. No computer is going to come to your rescue when you have to explain why you wrote the code you wrote. Encourage the use of ChatGPT and other tools. Put the emphasis on understanding what the code does and why. The tool may help to explain this to the student, but if so, that’s fantastic. No need to…

If you can’t actually write undergraduate level code then seeing a machine do it and all you can do is pore over it and try to explain it simply amounts to you knowing less and being less capable of computer science than the average student in the 70s or 80s. This isn’t like an assignment to take an existing Linux kernel module and explain it given the source code. These are basic primitives.
Post reply on HN