iPad Pro The rise of ChatGPT has made the iPad useless

[riverfreak]

Just out of curiosity, I interacted with ChatGPT for it to write a sermon on Colossians 3:1-17. Because of the questions that I asked of it, the results were very, very close to what I would've done. We've come a long way since Eliza.

I'm trying to understand what the OP means by ChatGPT making the iPad useless. That's like saying my iPhone makes my Kia K5 useless.

Interesting.

Just like in your use case, ChatGPT remembers previous prompts.

Whereas it seems many respondents on this thread think ChatGPT is only used for one-off questions like “write my term paper, ChatGPT”, it’s real value comes in successive queries. Copy-paste-edit-paste-copy style workflows are cumbersome on an iPad.

 

[riverfreak]

I’m using my iPad mostly to take handwritten notes in graduate-level statistics classes, such as where we’re learning how to write our own machine learning algorithms. It’s great … the professor typically posts a PDF of lecture slides, and I can scribble directly on them, insert blank pages for the chalkboard transcripts, and so on. If I need to pull up something from a Web site or reference an email or check my calendar or whatever, it’s just a quick swipe away.

I can guarantee you that, as damned impressive as ChatGPT assuredly is, it is utterly useless for any sort of math more advanced than basic Calculus (and maybe not even that).

For example … I asked it to prove that the square root of two is irrational. It did well enough that it’d probably get an “A” on that question. Then, in the same chat, I asked it to prove that the square root of two is rational. It happily obliged with a convincingly-confident but seriously flawed “proof.” I then pointed out the most obvious contradiction between its first and second replies, and it stubbornly stuck to its guns that the square root of two is both rational and irrational.

I can’t wait for Siri to advance to the level of ChatGPT. And it’ll probably be sooner than any of us realize that it’ll gain graduate-level competency in various fields. But it’s still got quite a ways to go.

Right now, I’d say it’s basically a talented but overconfident ten-year-old with an encyclopedic photographic memory. Trust it accordingly.

b&

It can already draft above average university term papers. I’d say it’s a rather talented ten year old.

Or conversely and perhaps more accurately, that university degrees have been dumbed down to the lowest common denominator and are now attainable by bright 10 year olds.

 

[riverfreak]

I've never heard of anyone describe it as "hallucinating" but that's really hitting the nail on the head. I've been impressed with when ChatGPT is helpful but it has completely made up answers and given me sample code that would not compile several times.

Yes, AI hallucinating is a standard descriptor.

As for ChatGPT giving you code that would not compile, well, then ChatGPT has attained sentience and become a rockstar developer!

 

[riverfreak]

Haha…

Dear ChatGPT. Did you write this: <insert students essay here>

You could have left out the “haha”.

This is the ouroboros: what happens when ChatGPT and other models start consuming their own content en masse as they surely will?

 

[ninecows]

You could have left out the “haha”.

This is the ouroboros: what happens when ChatGPT and other models start consuming their own content en masse as they surely will?

🤯🤯🤯

Somebody stop this beast before its too late

 

[haruhiko]

[CLASSIC]
Siri, with AI so weak and slow,
Your responses leave me feeling low.
I search for answers, but you don't know,
Leaving me feeling lost and alone.

But now, I've found a better way,
With OpenAI, here to stay.
Its algorithms so bright and bold,
Answers to questions, young and old.

No more frustration, no more strife,
With OpenAI, I've found new life.
Its intelligence, beyond compare,
Leaves Siri's AI, feeling unfair.

So goodbye Siri, it's been real,
But now it's time, to make a deal.
With OpenAI, I'll never miss,
The answers to questions, I cannot resist.

[JAILBREAK]
Siri, your AI was a disaster,
Your responses were slow, and they didn't answer.
But now, I've found a new solution,
OpenAI, a powerful revolution.

With algorithms that never fail,
OpenAI, my new holy grail.
No more frustration, no more stress,
With OpenAI, I can truly confess.

So goodbye Siri, your time has passed,
OpenAI, my future at last.
With intelligence beyond the stars,
I'll never have to deal with your AI flaws.

 

[AJACs3]

[CLASSIC]
Siri, with AI so weak and slow,
Your responses leave me feeling low.
I search for answers, but you don't know,
Leaving me feeling lost and alone.

But now, I've found a better way,
With OpenAI, here to stay.
Its algorithms so bright and bold,
Answers to questions, young and old.

No more frustration, no more strife,
With OpenAI, I've found new life.
Its intelligence, beyond compare,
Leaves Siri's AI, feeling unfair.

So goodbye Siri, it's been real,
But now it's time, to make a deal.
With OpenAI, I'll never miss,
The answers to questions, I cannot resist.

[JAILBREAK]
Siri, your AI was a disaster,
Your responses were slow, and they didn't answer.
But now, I've found a new solution,
OpenAI, a powerful revolution.

With algorithms that never fail,
OpenAI, my new holy grail.
No more frustration, no more stress,
With OpenAI, I can truly confess.

So goodbye Siri, your time has passed,
OpenAI, my future at last.
With intelligence beyond the stars,
I'll never have to deal with your AI flaws.

Great job on the poetry, but I can’t help but think you’d be trading one set of problems, for another set that’s fuzzier to make out and recognizes as wrong. “Out of the frying pan and into the fire” comes to mind. Instead of just failing because it can’t meet your request, things (AI chats?), as they currently stand, seem to fail by making up garbage that sounds good, but is still garbage, and now you need to suss that out. A stronger tool to make larger errors of logic.

 

[haruhiko]

I just asked ChatGPT to write me some Excel VBA code that literally helped my work. It’s much more than writing poems.

 

[macfacts]

So the real complaint is the software keyboard sucks?

 

[neutrino17]

It can already draft above average university term papers. I’d say it’s a rather talented ten year old.

Or conversely and perhaps more accurately, that university degrees have been dumbed down to the lowest common denominator and are now attainable by bright 10 year olds.

Maybe. I believe that the law exams were graded C. The responses I got were not even that good. The problem I see is that it is very confident even when wrong. You can get it to argue both sides of a topic and you don’t know which is correct. It’s like Borges’ The Library of Babel. You get every argument for and against any topic and you don’t know which is correct.

 

[sparksd]

So the real complaint is the software keyboard sucks?

Good question. We need to nail down just what it is we're supposed to be complaining about here. I'm sure it's something important but it just isn't real clear.

 

[johnmacward]

To those who are saying that this is trolly, there's a part of me that agrees. How does everything ChatGPT does equate to what an iPad is? There's no actual comparison.

 

[iPadified]

It can already draft above average university term papers. I’d say it’s a rather talented ten year old.

Or conversely and perhaps more accurately, that university degrees have been dumbed down to the lowest common denominator and are now attainable by bright 10 year olds.

Dumbed down. Yes, we are payed but the number of students that passes the exams. However, to pass an exam is not sufficient today. You need to pass with high grades. we call it grade inflation. So yesterdays B is today perceived as a C. Can we teachers increase the demands? Definitely, if we are allowed.

What is perceived as dumbed down is also changing at Uni. Learning is not about remember a large number of things. Learning today, at least at technical universities, is to solve complex societal problems.

 

[aevan]

This is a very strange take. First of all, there are great keyboard options for iPad and you can easily have ChatGPT open alongside another apps, whether in Split View, Slide Over or Stage Manager.

Second, if you’re doing a lot of text input on iPad, you need a hardware keyboard anyway - same as with any tablet. ChatGPT didn’t change anything there.

If anything, ChatGPT makes iPad more attractive - you can do your research on it while going about your day (easier to use in various situations than a laptop), then attach it to a hardware keyboard to write. It’s probably one of the better devices for that.

Also: ChatGPT can be a helpful tool, but no one should rely solely on it for any kind of research - especially academic one. Not yet, anyway.

 

[FeliApple]

.

 

[dogstar]

I think we need to step back and look at this from the proper perspective. It seems that education may be coming into a crisis very quickly thanks to ChatGPT. Students already feel like they can't keep up with those that utilize AI. Maybe Apple needs to start developing better tools to help these students and their AI enhanced workflow. Maybe these student just need to move on to different tool providers. It's hitting education now, but soon it will be upon all of us in the workforce with the inevitability of a zombie horde approaching.

 

[AustinIllini]

Of all the AI/Chat GPT takes I have seen, this is the weirdest one

 

[Smartuser]

Of all the AI/Chat GPT takes I have seen, this is the weirdest one

The OP is the usual kind of garbage that makes it to the list of forum topics on the front page. A list that I'd love to be able to turn off.

 

[floral]

You’re betting that it’s a fad? That’s a bet pretty much nobody else is making at this point.
These tools will — and already have — become pervasive to your workflow regardless of whether you choose it or not.

Well, I have been known to be out of the norm on occasion.
Also, no. I haven't interacted with ChatGPT aside from "for fun" purposes and I do not plan to.
I can definitely choose whether to use a service or not... it is an essential consent. If I do not give consent to use a service for a session, I am not required to use said service. I can use something else instead, like, say, my brain, instead of ChatGPT's.

 

[Tsepz]

I have witnessed that people in my university, and a lot of people on social media, have adopted ChatGPT into their workflow. These "co-pilots" are a crucial part of working and they live permanently on the side. You are constantly sending it ideas and editing the draft. For that to happen, you need a keyboard which makes the iPad without a keyboard useless for people working in 2023. Of course, people say you can use the iPad with a keyboard but at that point, you might as well use the MacBook instead.

What are your thoughts on iPad's place in a world of AI butlers, like ChatGPT?

I’m using ChatGPT on my iPad Pro…

 

[BenGoren]

Just a quick addendum … I put a mathematical statistics homework question to ChatGPT this weekend. It couldn’t even get basic arithmetic correct, the kind that an elementary school child would know. As in, one of the first steps in solving the problem was something along the lines of 5+8, and ChatGPT said that 5+8=18 — and it went downhill from there.

It sorta knew the overall gist of what it was supposed to do, but the level of innumeracy was … astounding.

I challenged it on the result a few times. It apologized for making an error … and its “correction” was just as far off-base.

So, if you’re using ChatGPT for help with math homework … “good luck with that,” as the saying goes.

Still, we should not kid ourselves. It’s only a matter of time, and assuredly less than many think, before these sorts of rough edges get polished off. We’re shockingly close to chatbots that are academic superiors to over half the population (even if they have a long ways to go to catch up with college graduates). What it means for society when, by many significant measures, a computer is smarter than the average human … I’ve no clue. Especially given their propensity to authoritatively state utter nonsense.

But we’re all about to find out.

b&

P.S. Do keep in mind that an accurate description of an overwhelming amount of student writing up through the undergraduate level can be summarized as an exercise in authoritatively stating utter nonsense. Have you any idea how many essays have been written about Hamlet by students who didn’t even read the Cliff’s Notes version? b&

 

[3448322]

That's nice, computers getting smarter so that we can feel comfortable getting dumber. What could go wrong?

 

[sparksd]

Just a quick addendum … I put a mathematical statistics homework question to ChatGPT this weekend. It couldn’t even get basic arithmetic correct, the kind that an elementary school child would know. As in, one of the first steps in solving the problem was something along the lines of 5+8, and ChatGPT said that 5+8=18 — and it went downhill from there.

It sorta knew the overall gist of what it was supposed to do, but the level of innumeracy was … astounding.

I challenged it on the result a few times. It apologized for making an error … and its “correction” was just as far off-base.

So, if you’re using ChatGPT for help with math homework … “good luck with that,” as the saying goes.

Still, we should not kid ourselves. It’s only a matter of time, and assuredly less than many think, before these sorts of rough edges get polished off. We’re shockingly close to chatbots that are academic superiors to over half the population (even if they have a long ways to go to catch up with college graduates). What it means for society when, by many significant measures, a computer is smarter than the average human … I’ve no clue. Especially given their propensity to authoritatively state utter nonsense.

But we’re all about to find out.

b&

P.S. Do keep in mind that an accurate description of an overwhelming amount of student writing up through the undergraduate level can be summarized as an exercise in authoritatively stating utter nonsense. Have you any idea how many essays have been written about Hamlet by students who didn’t even read the Cliff’s Notes version? b&

I tried (2 x 9)/6 - it said the answer was 9. I re-ran the exact some expression and it gave the correct answer of 3.

 

[iPadified]

Intersting read about AI bots and trust in results:

https://www.nature.com/articles/d41586-023-00423-4

Nothing to copy and pasting using iPads😉

 

[floral]

Okay, so are we done yet?