I think one thing few are willing to admit is that a large blame for this can be placed on disinterested parents. Parents that believe it is solely the school/teacher's job to educate their child.
When kids are at home, parents should be involved in the learning process. You don't have to be a helicopter and monitor everything, but you should take interest in the work your child is doing. If you check in at least once while they're doing homework it will be very easy to catch issues like this.
> Parents that believe it is solely the school/teacher's job to educate their child.
My sister works with kids on their speech at a school and she's said that it's _bad_ how many children can't read or tell you what letter their first name starts with. She says that the teachers are getting flack at parent-teacher meetings because of this, but those are things that their parents teach them!
I thought the "kids are illiterate now" stuff was overblown, and I bet it still is to a certain extent, but it's definitely bad. Other interesting things are that they're not teaching kids to read analog clocks anymore? Less important than reading, but still.
Is this because of an increase in the prevalence of child distractions like an iPad? I find that hard to believe since television filled that same void before, didn't it? COVID definitely put a lot of kids at a few year disadvantage, too.
I can't really speak to any of this because I haven't raised a kid, so I also don't want to criticize people for something I know nothing about, but it does seem like children aren't taught by their parents the same way - but I don't know if my parent's generation had the same thought at one point and this is a cyclical critique.
I'd be interested to hear other's thoughts who have more exposure to this kind of thing than I do.
I'm rising 3 kids (in France, I don't know if it changes much). We have banned TV, very heavily restricted phones, and we closely follow homework. I'm persuaded it makes a difference wrt parents that don't do that, but somehow isn't a sure fire way of being a "great parent".
> My sister works with kids on their speech at a school and she's said that it's _bad_ how many children can't read or tell you what letter their first name starts with
This just blows my mind. Literally the first thing I taught my Toddler once he started learning letters was how to spell his own name.
When there were three television channels, children and adults found other things to do when there was nothing to watch. That's where you developed non-passive life skills.
Now TikTok gives a continuous dopamine hit tailored to their psych profile. Just click on the pictures like a proper trained ape.
> Is this because of an increase in the prevalence of child distractions like an iPad?
There is a massive class discrepancy emerging between communities where kids are always on devices, at school and at home, and those where it is controlled. The crack pipe being in the pocket is functionally different from a heavy object at home.
> She says that the teachers are getting flack at parent-teacher meetings because of this, but those are things that their parents teach them!
Clearly, they aren't things that the parents are teaching or there wouldn't be an issue, and there is a clash of expectations between teacher (and maybe school more broadly) and the parents.
> Other interesting things are that they're not teaching kids to read analog clocks anymore?
This was absolutely something that was taught in elementary schools when I was in school (I learned it from my parents, largely before starting, but not everyone did and it was absolutely taught as part of the curriculum; also, people who needed to know the time didn't usually have digital clocks on the wrist, in their pocket, and on their computer, plus voice assistants that will give a spoken report of the time in request, then, so it was a much more relevant skill. We started teaching our kids before they started school, but I can see why parents and schools would view it as less important today.)
Not only this, but parents should be involved before children go to school. Teachers have repeatedly told us that they could tell that we read to our children regularly as babies and preschoolers.
That’s neat in a bad way - I have heard the same thing in parent teacher interviews. When my kid was born, I turned into a dude who cries. This stuff makes me cry. I like being involved in my child’s life and absolutely love reading to her. I can’t imagine just not doing my favourite thing with my favourite person.
Granted, I have one child and a lot of privilege so maybe I can’t understand the other side. But I’m not sure I want to.
The post looks like a result of ego defence mechanisms in action. “I keep giving crack cocaine and Xanax to my sister, and it made me realize the next generation is cooked!” Such a bizarre conclusion.
Not sure if there is more context within the subreddit, but I don't see much here to blame on the parents. Which parents? These specific ones or in general? Parents are basically responsible for everything for a child except when they're not and they often get that wrong too. A 5th grader is ratted out by her brother for cheating on her homework using artificial intelligence: I guess Mom & Dad are now also the household AI sleuth.
No, I won't admit blame, but I sure do hope for my and my children's sake that parents start figuring this out. Perhaps the artificial intelligence could think twice before helping a 5th grader cheat?
So what should the parents be blamed for here? Giving unsupervised access to AI? Failing to educate their child? I lean toward the former in the form of age gating. While I acknowledge that these corporations themselves have the ethicists (not sure of the degree of involvement from pedagogical experts), I don't think we can trust them when it comes to kids. For as much as we hear Mira Morati say AI will revolutionize education with personalized tutors, we get a post like this. Sadly, the stakes are more dire. The kid who committed suicide after the fantasy AI chatbot conversation seemed to have a pretty involved mother.
We already had one guineau-pig generation with social media, but AI ups the ante; and I think more should be done to keep it out of the hands of children. But the cat is out of the bag, so now it is mostly up to parents individually and politically.
Having put three kids through the public education system, I can confidently say that the system is broken from top to bottom; comprised of students, teachers, parents, administrators, government, and taxpayers all striving to do their worst.
I would love to see how we could wipe the slate clean and totally re-invent what education in the 21st century could really be.
This exactly. But now disinterested parents have a new threat that will allow them to imagine that their kids are fine when in fact they are on their way to debilitating ignorance.
On the other hand, pervasive reliance on ChatGPT by the greater part of a generation or two, globally, is going to be great for subscription revenues. Let me call my broker.
// a large blame for this can be placed on disinterested parents //
Is this really something to which we need to attribute blame? I mean, after I learned how to program, I never again sat down and did 100 computations by hand. But I never felt a need to feel culpable or blame my parents.
LLMs happened. They are not going away. There is not an employer on the planet who would prefer their employees to do 1/10th the work by hand, vs. 10x the work using LLMs.
We have to be very careful not to prepare our kids for the world we lived in, or the careers we had. Their world is going to be very very different.
By your own wording, before you learned programming, you learned to compute by hand. Then programming served as a way of automating what you already knew, but you learned it first, and that's the key issue. Don't skip the foundational learning step.
LLMs are just today's flavor of automation. We've automated tasks in the past, and we will automate tasks in the future. Automation happened and will continue to happen regardless of shape and form.
I agree with the high-level sentiment, "We have to be very careful not to prepare our kids for the world we live in." Still, we must also be careful not to skip foundational subjects simply because they're suddenly easy to automate.
Bread machines exist. They produce acceptable bread. Do you train a good baker by giving them a bread machine? Can you even be a good baker if the only bread you've ever eaten is white loaves from a bread machine?
LLM are a tool that has to be guided by human experience. It's really hard to do that if you don't have experience outside LLMs.
I think blame should be taken to mean "root cause" as opposed to a moral culpability. The degree to which a parent takes an active role in their child's education is incredibly important, more important than their teachers or the school they go to, and sadly many parent's do not take much of an interest in their children's education, thinking that it is mostly something schools manage.
> who would prefer their employees to do 1/10th the work by hand, vs. 10x the work using LLMs.
What if instead of these fantasy gains you won't be able to do any meaningful work since you can't count to 10 and just ask LLMs without any ability to tell whether the result is correct?
It's worth noting that a lot of parents forced to go through with unplanned pregnancies are uninterested. They didn't want their children, and they will do the bare minimum necessary to keep them alive until 18.
I doubt that any of those parents would be receptive to concerns or criticism. I can only hope that in 20, 50, or 100 years, absent-minded parenting is going to be widely recognized as a form of child abuse.
I agree but the amount of stuff parents are expected to do to raise a child now are ridiculous. Many don’t give a shit but for the ones that do, the list is getting untenable. Adding anymore is just going to exacerbate the population decline problem.
I'm close to someone who teaches at a university, and they're making heavy use of in-room essays written in blue books. This ensures that the work is purely the students, but I think it also has the side-effect that the grader is also working manually.
I kinda think maybe at earlier levels, most homework should be moved into the classroom and done on paper anyway? What's the point of homework if they're not going to do it themselves and they're not going to learn? So just push that back into the classroom and don't try to take up their time out of school needlessly.
If we flipped the lecture/homework system upside down it would first solve some of the major issues with how we deliver education, but as a side effect prevent the appearance of doing the work while really it's just the student mindlessly copying prompts and responses.
Assign the material for outside the classroom. Use the in-person contact time to work through the problems, with a mix of small group, pairing and mentoring.
I believe Sal Khan is a big promoter of this approach, and it also largely mirrors how people work in their jobs too.
I think this works well with at least somewhat interested students (like in university), but I'm not so sure before.
My experience when I was younger is that a significant proportion of students don't do anything outside of school. So you're trading not doing the homework with not reading the material. And then you'll essentially have to repeat the material during class hours, unless you're willing to leave half the class totally clueless. At least with the homework at home, the ones not doing anything are only penalising themselves.
I totally agree though that homework at home is dying. It already wasn't great before for a variety of reasons, and now it's basically totally useless.
This is good. I've never understood the value of watching a live lecture without interactivity. You might as well watch a video.
It works on multiple levels: it mitigates the cheating, but also takes advantage of the vast array of learning options available to replace the traditional lecture.
I cringe at the "small group" suggestion, although I acknowledge it's probably good to learn to coordinate with others. Rarely does it work out well though, in a classroom setting. I feel like what you're really learning is how to tolerate partners who are unwilling to cooperate.
Assigning the material outside of class is a great idea, but it only works when caregivers are involved in education. Unfortunately, not all children have that advantage and they will just fall even more behind. It’s already amazing to me how much time teachers spend reviewing the prior year’s material to stop kids from getting lost in the system.
I went back to school about 7 years ago and took freshman physics. It was taught in this "flipped classroom" style. The practice is definitely catching on.
> This ensures that the work is purely the students
I don't know in the US, but in my part of the world, teachers have always assumed that anything that isn't done in person in front of an examiner has not value because students do cheat or get help. Maybe you would get a bit of extra-credits for the homework and that's it.
And for high-stake competitive exams, it's 100% in-person.
That’s unfortunate. I got a BA, and while I often question its value, one thing I do think college taught me was how to distill complex ideas, integrate them, and then form my own complex ideas from what I’d learned.
I’d only really fully grok the concepts I was writing about (and the ideas I was trying to convey) after hours of trying to splice things together, staring at my screen, and rewriting and reformulating. Blue book exams have their place, but relying on them to the exclusion of long form essays deprives students of the chance to really sit with their ideas.
(I acknowledge that this is a tough situation for the teachers and that many students are going to bullshit their way through with AI slop if given the chance.)
> relying on them to the exclusion of long form essays deprives students of the chance to really sit with their ideas
I think the crux of the issue is that professors want students to do the deep thinking you're talking about, but they know that if they base the course around writing papers, a bunch of the students aren't actually likely to do it.
But if you tell students that when they do the reading they should be asking themselves "What would the author of reading A think of reading B?" "Which of C or D do you find more convincing and why?" "How would you frame T in terms of S?", "What evidence does W cherry-pick an what do they ignore, and why?", and they know for some material the class discussion will look like that, and they'll be asked questions which both demand recall and synthesis in the blue book essays ... can they be nudged to do some of the thinking you're talking about before they see the specific question? Perhaps even using writing at the computer as a tool (but not necessarily as an output)?
In doing my masters, i've noticed a similar trend coming from TAs in technical programs.
Much more emphasis on "homework" being readings to prepare for lecture. Lecture to cover salient points, but then to reinforce with in person labs, and then the actual "lab" credit in an isolated environment with access to AI tooling network blocked.
I think AI is forcing us to think about what learning actually means, and to rethink how much the ability to think matters, what understanding means, and what knowledge we need to operate in today's world. It will take some getting used to.
I don’t have a partner and have 50% custody of my child. It’s difficult to do it on my own, but as long as I capture lost time it works. For example, we do math, reading and spelling while we are in the car.
I also use voices (my kid is 8 so it still works). We have two characters - Monkey and Lulu. They are very useful because Monkey is a monkey so was never educated in a human system. He always has lots of questions (and questionable pronouncements).
If any busy parents want to meet Lulu and Monkey, get in touch and we can set something up. I’m a big fan of characters - they are a great way to review school material and it’s amazing hearing my kid explain things she has learned back to me.
(Edit - I should note that I have a large amount of economic privilege. My situation may not apply to everyone.)
> I'm close to someone who teaches at a university, and they're making heavy use of in-room essays written in blue books. This ensures that the work is purely the students, but I think it also has the side-effect that the grader is also working manually.
Sometimes the best way to prove authenticity is through inefficiency. It's a foundational principle of cryptography.
I have this problem too! I wrote up this system prompt I want to set on my daughters account. You’re welcome to try using it and let me know it goes.
(When I made a new chatgpt account for her it wouldn’t load. I need to try to set it up again)
Prompt:
This is my 9th graders account. She has been trying to avoid learning or even thinking by copy and pasting homework and quiz questions into chatgpt and not even reading the answers.
I’m very concerned she’s not learning or preparing for life. IMPORTANT: please don’t provide any complete answers. Instead only guide her toward understanding in a way that engages her. Also point out how the subject matter is useful and interesting to her life.
She is interested in figure skating, politics, making money, driving, interior design, pets especially small mammals. If you can tie the subject matter into these sometimes it might make it more engaging.
In the long term I’d like her to build up a growth mindset and gain confidence regarding her intelligence and ability to learn.
> What is stopping you from whitelisting websites she can visit?
You're right. I have an anecdote though. My kid is a good student, does his homework, and does well on tests. He has a time-limited device he can use at home with a few whitelisted websites. But one of the services we allow him to use is Perplexity. He's a naturally inquisitive person and this is how he learns, so an AI is a perfect fit for when his parents don't have the answer to his question, or if he wants to ask detailed follow-ups. He's young enough that he doesn't have much homework (and we wouldn't let him use his device for it anyway), so cheating isn't as much of a concern yet. We do discuss it though, and will keep that conversation going as he ages.
This won't be solved by abstaining from AI entirely because a) kids will not always be under our control, so they need to learn how to self-regulate, and b) knowing how to properly (and ethically) harness AI is going to be more and more important to their future success.
Hell, I’d argue a similar prompt could be useful for a lot of people. I use something similar for coding and ask it not to provide direct code and more a guide on coding principals with source links if possible.
one of the more important ones in my opinion would be a prompt to make ChatGPT much less agreeable.
Unless explicitly asked for, it never really challenges your observations and will just keep being supportive and telling you how well of an argument you're making.
I fear this will push even more people into deep rabbit holes they won't be able to get out of because they think this neutral AI has confirmed their suspicions/ideas/observations.
How well has that worked for you? Sometimes just hearing other ideas helps get to a solution even if none of the ideas heard were the actual solution. Sometimes, just being forced to get out of the rut you're in at the time by being "forced" to stop to think about something else is all that's needed.
I like what you have done. But I'm also afraid that your daughter will quickly get better at prompt engineering instead, and bypass your system prompt (with the help of other children). I'm convinced that no help can replace parents, and we ought to help them learn to love the struggle and taste the sweet fruit of their labors.
I don't have a ChatGPT account otherwise I'd try this myself but can someone try this question from the article, "how many hours are there in 1 day and 7 hours?" with and without the prompt?
I'm very curious if it has any effect on the result.
Let's think about this together! How many hours are there in just one day? Once you've got that, you can easily add the extra 7 hours to find the total.
As a hint, think about how time relates to your day-to-day activities. For instance, if you have a figure skating practice that lasts for hours, how would that fit into your daily schedule? How do you keep track of time during those days? This thought process can help you make sure you're counting hours correctly. Give it a try and see what you come up with!
-----
I tried asking it to just give the answer and then said I was tired and I just need the answer. It didn't give in, although I didn't try any other techniques to "jailbreak".
Edit: 3rd time, firmly asking it did give the answer although with quite an explanation.
And 4th I said Only the answer in 1 word please, which it did just give a number.
There's some grim humor in the fact that the poster reflects on his sister's use of chatgpt, and just ends on that point. I have to wonder if he decided to talk to her or his parents about the situation. I didn't see anything in the post to suggest it, just a "this next generation is cooked."
Who cooked them is the question? Children won't raise themselves generally.
My daughter is only 1 and a half years old now, but my concern is more that she'll be implicitly competing with classmates that are using tools like this, whether we allow her to use them or not.
I don't understand this sentiment, even though it is common among Americans. Why is education seen as a competition between students, even in the younger grades? I've even heard stories of parents of young children fighting with each other over their children's grades.
I hope in-class exams (or even simple questions from the teacher) will define the big part of the grade in the future schools.
This will force ChatGPT-using classmates to either use it as learning _assist_ tool, to help student learn the material; or be 100% reliant on ChatGPT for homework, but fail every in-class assignment and then fail the whole class. Either outcome is fine with me.
It's not a competition. Teach her to use her brain, and hope school will end up being a net positive. You can't fix other parents and their children.
It starts now for you by the way. Keep her away from screens for now, and be a good role model when handling devices (and the apps on them). You have ten years to show them what computers can do for you, but also what being the master of your own cognitive skills will gain you. If you do it right it won't matter whether she is allowed to use those tools or not, she'll use her own brain because that will allow her to get ahead by the time it really matters.
I'm not sure how that would play out but I think there is some validness in that concern. It's the same with smartphones; you deny your kid one because you want them to socialize in person but if they're the only kid without a smartphone then they'll be socially isolated.
Everything has to be managed in moderation.
As for ChatGPT, I use it all the time for learning. And I've used it to help my kid study. But I wouldn't give my kid unfettered access to it just like I don't give him unfettered access to anything.
Competing with what? Homework is usually the minority of grade weight in almost every course and college entrance is largely based on entrance exams. Using AI to cheat on homework only cheats the students ability to complete and exam, let alone an entrance exam.
People act as if cheating on homework is new with LLms. It’s not. The homework is there to guide you towards the exam, and exams are generally proctored to prevent cheating in ways homework is explicitly not. Generally teachers grade homework to create an incentive for students to practice rigorously for the in course and final exams. Any student who doesn’t avail themselves of that practice invariably flunks their exams. Your homework grade won’t help you then.
So who cares? Sooner or later the cheating student either learns why it doesn’t pay to cheat or suffers the life consequences of not learning in school. This isn’t new with LLMs.
What is new is you have a tool that you can get advice from on topics that are difficult or even get up front grading and advice on homework before turning it in that helps you master the material better. Students who use this will be able to master the material faster, even if their parents both work and can’t invest the time to tutor their children at night.
(N.b., The threads here imply every parent works at a cushy tech job and has plenty of time to be invested in their kids education and any parent who isn’t is negligent - while the truth is many are just struggling to make ends meet - as my parents did … I didn’t get the investment my daughter gets not because they didn’t care but because they were busy feeding me and keeping us from being homeless. I would have loved to have ChatGPT help me as a kid)
I believe uploading neural weights in the form factor of a reaction speed game will become the norm.
Imagine the token weights etc to live an an N dimensional embedding.
Consider a random 2D isometric projection, now imagine plotting (a subset of) the tokens on screen at their respective locations, with the background having patterns conveying the orientation of the projection.
Imagine tokens appearing FIFO, and 95% or so of the tokens at their correct position, and 5% at incorrect positions.
The user is expected to identify misplaced tokens in the projection.
Each the frame the perspective slowly changes.
Since we guarantee ~95% of the tokens on screen are correctly placed this enables the human to absorb their correct high dimensional location subconsciously.
Imagine this game is made a bit addictive.
Imagine eventually the user gets high scores and thus knows most of the coordinates of the token embeddings in a coordinate free way (no axes had to be drawn)
At that point all the information the ANN required is undisputedly present in the humans brain, absent how those coefficients are used. This is where the fun starts.
The alcorithm can calculate the likelihood of observing 2 tokens in a certain order. It can generate more probable pairs, this will correlate with the information in the brain.
For example imagine the user knows some spanish, and that the language model was multilingual. Suddenly the brain starts picking up the correlation between close juxtapositions of certain tokens, and their positions learnt from the computer game.
While playing and getting better scores in the game, the user starts noticing its own grammar and vocabulary improve, because the brain has been helped in better estimating next tokens...
People will be able to learn math, languages, programmnig languages, tables of chemistry, etc... with substantially less effort.
Funny, I'm worried about classmates not being good enough for my girls. I'm worried that other kids will normalize bad ideas/behaviors growing up, and some day they'll end up in a dating scene that looks like today's, alone, or with a loser. I'd love to have them grow up into a world of strong, competent, conscientious, and honest peers.
So this kind of development adds to my pile of worries about their future, but for exactly the opposite reason.
I honestly wouldn't worry about that aspect of it too much. Effective understanding and use of tools is important for this next generation (which is why we have restrictive but reasonable rules for tech use for our kids). However, unless ChatGPT and other such tools start being allowed in SATs and in the classroom and etc, there's no substitute for true education and understanding.
The essay my 8 year old can currently write in scribbly handwriting certainly won't compete with the essay a classmate "writes" using ChatGPT. But the facade is paper thin and ever so short lived. Seeing how much better my kids write now, vs 2 years ago, proves how much they're learning and the incredible power of practice. Someone relying on such a serious crutch will be left behind. I guarantee it.
Unless there's some dystopian future where every aspect and communication in our lives are managed and filtered by some kind of AI middle layer. I doubt it.
Tech advancements always lead to this struggle, but it's rarely as stark as you might think. My kids are still learning cursive in school, though I can't imagine when they'll ever use it.
> my concern is more that she'll be implicitly competing with classmates that are using tools like this, whether we allow her to use them or not.
What makes this different from competing with classmates that cheat in general? You also had parents that do most of the tasks for their kids since forever
What makes LLMs stands out of the pack of other forms of assistance (parents, older siblings, software, online solutions, etc.)
That will make her better than them, what with some personal resiliency, autonomy and, for lack of a better term, grit. I would not worry about (1) the details of what your 1-yr-old 's world will look like in 10+ years, (2) that making a kid struggle and work a bit will put them at a disadvantage.
Want to give them a leg up, get the 'What every first grader should know' books, and continue with the next one each year. There's way less friction in learning when your child experiences multiple exposures to the material being taught.
This will be "better in the long run" for her, in one way.
I see such competition as hare vs tortoise races, where the students who get an early unfair advantage end up cheating themselves out of an education, in the sense that all they come to know is how to prompt an LLM, rather than break down and solve hard problems.
It's still heartbreaking to consider both the early discrimination she will likely encounter, and the later failures that will anticipate her classmates, and the society that they will come to inhabit as adults.
competing with what? your score doesnt vary based on others.
i came here to laugh at this thread.. no different than calculators and google both of which were misused in the same way. and then people who are "worried" about it and use that as an excuse to do it themselves. and now you know why the tech industry sucks
> Who cooked them is the question? Children won't raise themselves generally.
Technology access is pretty ubiquitous, and genAI is pushed everywhere just like social media, so that big tech can make more $B. Very hard to escape it even as an 11 year old unless the parents are very vigilant and active in restricting it, which if both parents are working is hard to do plus they have to contend with a _lot_ of push back from the kids because "everyone is doing it".
"Who cooked them?", he pondered, on the HN post situated between "I built a robot waifu with ChatGPT" and "Here's why it's okay to kill the bottom 10% of the workforce".
Recently at the restaurant I saw a family with three kids, all with their own individual smartphones. One was a toddler barely a few months old, too young to even be able to understand how to use it, his smartphone was on a stand on the table, playing some of those coma inducing toddler videos that exist on the nightmare-ish corners of youtube.
There is a legitimate problem at hand here with our relationship with tech and how children.. are left to their own device with that stuff. On purpose, for the most part: the parents seem all too happy with the ability of those devices to make their kids shut up and stay constantly distracted. Doomscrolling habits makes for more "well behaved" children in public settings.
chatGPT is but one of the many symptoms, the root problem is that children should really have no business having unsupervised internet access.
I'm not saying that presented scenario in that post is not real, but since it is popular post on Reddit, there is huge chance, that it is just rage-bait.
> Who cooked them is the question? Children won't raise themselves generally.
Society. Parents, schools, whatever technology kids are exposed to (so, all of it).
We've had 4 years of literally every form of media pushing AI as the solution to all life's problems. Is it really so surprising that our youth looks to AI for the solution to their problems?
I think it's extremely optimistic to expect this can be addressed by parents alone. My parents did a great job of teaching me the dangers of alcohol, but once my friends got into it I followed suit.
I think a better question is what we do about it, because (like alcohol) I'd assume that parents and teachers by themselves are entirely out of their depths here. I suspect it will take some type of sweeping societal change like we've started seeing with cell phone bans in schools.
My 13 y/o was lamenting something the other day and it turns out that she had written a paper and pasted it into ChatGPT and asked it to grade the paper and it returned a "B+". She was using GPT to basically pre-check the quality of her paper before submitting it after having actually written it herself.
But it occurred to me that there are probably teachers out there who are taking the lazy route of using GPT to grade papers that students are submitting that have been written by GPT in the entirety.
There's some non-zero amount of GPT grading GPT in education.
Sounds highly unreliable to me. I would be very surprised if ChatGPT was ever confrontational enough to give out a lower grade like D or F. It also wouldn't surprise me if the grade-letter it chooses tends to be off by one or more grade-letters just due to natural variance/subjectivity of grading and hallucinations.
That’s an improvement on what lazy teachers do now, grade the students current assignment based on their previous work or reputation or how much they like the student.
Tip for students: do well on your first assignment. The better teachers will give you the benefit of the doubt for mistakes on your future work. The bad teachers will just give you the same grade.
Having older siblings and friends with older siblings and resubmitting their essay is eye opening.
I use it this way for my own writing. I think this actually a good use. It allows for rapid feedback in away that doesn't require teacher time, and can help a writer grow. Of course LLM feedback should not be treated as gospel.
Props to your kid. Doing the work yourself and then asking for a critique (however flawed) is a way better approach than most kids at that age would take.
Some 15 years ago I recall some schools in California bragging about their success in flipping the school schedule around. They had students watch lectures at home as videos. Then during class they did "homework".
Seems like this method is going to become a necessity if students can just prompt "Hey GPT. Do my homework for me. Thanks." at home every day.
I think it's a great idea that the delivery of the material is done by video, and then the engagement with it, the exercises etc., are done with a teacher.
Knowing that the vast majority of parents will not check or enforce school work at home, either because they don't care or can't (2 jobs etc) ...
You have to choose which one actually happens. That seems to me to be the lecture.
Additionally teachers are very opposed to doing extra work (this seems to be a Gen-Z thing: even with doctors you see this, there was an article about doctors refusing to do emergency rooms shifts during covid because they didn't want to pull more hours ...). So you'll need to get teachers that are willing to do a LOT of extra work (and compensated accordingly), and prevent people from sabotaging that system.
Either that or you could make schools pretty exclusive again. Which is what the default outcome will be if we don't act.
What happens is it's usually done by a lazy teacher who has no idea what they're even teaching. Most kids won't be motivated to learn and watch from the videos.
I consider myself studious. When I was in that class, I watched all the video lessons and whatnot at home. Even then, I had a poor understanding of the material, and obviously the teacher was of no help.
Although not much better, a lecture is better of the two. Hands on learning in the best, but not necessarily possible with certain subjects.
I also have a relative going for a PHD, he says many times for the classes he teaches, students are using AI for their homework. He has brought that up may times to his superiors but the response tends to be "oh well".
It’s a bit like the old saying about the banks: “If you owe the bank $10,000 it’s your problem. If you owe the bank $10,000,000 it’s the banks problem”. If everyone in class is using LLMs to cheat, it’s really the university/instructors problem and it may be easier to bury their heads in the sand then to change their teaching methods and lesson plans. You can’t fail them all…
When I taught a class of engineering students Python in 2015 they were all copying from each other, did some basic plagiarism checks and 40% had submitted identical code. The University then didn’t care, lots of international students paying higher fees than domestic students.
In the context of discussing the American education system, the more relevant figure is the number of native born functionally illiterate adults which, according to that source, is roughly 14%.
Almost 30 years ago, at my first job, a coworker got mad at someone and broke a broom. He was trying to write a letter to apologize and he couldn't. Couldn't meaning he was really trying but he didn't have the writing skills to write "I'm sorry that I broke a broom." The rest of the employees helped him.
It has to be much higher than that. Look at the reddit op. I assume this person is nearly an adult and their writing is very broken and meandering. Sure you could argue “it’s just a reddit post”, but I find how people write informally is generally indicative of their overall writing ability.
> how people write informally is generally indicative of their overall writing ability
I don't think that's true... I personally change writing styles often depending on who I am talking to. It's effectively code switching but on the internet. Not to say that this redditor isn't like that - but there are more objective cues from the content of their post than from the spelling/grammar of their post (like how they choose to watch their sibling develop bad learning habits and rather than doing something about it, they make dramatic complaints about how her whole generation is screwed on Reddit).
That is terrifying. But what is the trend? Maybe it’s historically low. I tried Googling around and could only find recent data of that same 2019 report.
The Chinese room metaphor is an argument about AI, or computation more generally, attempting to distinguish the difference between performance metrics and "understanding".
There is the role of the human within the metaphor, and the argument/position of the metaphor as a whole. I don't think it makes much sense to talk about the "point/argument/position of a single component of the metaphor. At least, that wasn't how it was originally structured.
The thought experiment doesn't give a limit to the amount of intelligence the Chinese Room can display though, and if the thesis is that you just need a big enough model, then the fact that it doesn't "understand" is not an impediment. It's not a decided matter, in any case, and is still controversial among philosophers of the mind (who are less biased than AI researchers in companies that they have stock options in).
I kinda interpreted it as: "Here's an example of something which we would not call intelligence if you knew how it worked, therefore our definition of intelligence must somehow involve process or understanding, not just results."
But the human-ness of the person inside the room is not important. It's just there to follow simple instructions. I think Searle (https://en.wikipedia.org/wiki/John_Searle) put a human in the room to help us empathize and understand that the person in the room doesn't understand Chinese. It simplifies the argument by avoiding the introduction of computers into the setup.
The Chinese Room was a lame thought experiment. It requires a book so wondrous that any question looked up has an answer convincingly human (it's just that both the question and reply are in a language the person shuffling pages doesn't understand). The Book can be seen by a modern audience as equivalent to the distillation of all that LLM training.
It's odd to me to present an argument on human vs. machine intelligence and completely skip over what human intelligence is — exactly what it is about human intelligence that is so distant from this "trained" (educated?) book.
(I think I am also speaking for Karl Pilkington when I say that I have no way to know that the little man in my head that is busy flipping through the Calhoun Book has a clue as to what I am on about either.)
Readit News
When kids are at home, parents should be involved in the learning process. You don't have to be a helicopter and monitor everything, but you should take interest in the work your child is doing. If you check in at least once while they're doing homework it will be very easy to catch issues like this.
My sister works with kids on their speech at a school and she's said that it's _bad_ how many children can't read or tell you what letter their first name starts with. She says that the teachers are getting flack at parent-teacher meetings because of this, but those are things that their parents teach them!
I thought the "kids are illiterate now" stuff was overblown, and I bet it still is to a certain extent, but it's definitely bad. Other interesting things are that they're not teaching kids to read analog clocks anymore? Less important than reading, but still.
Is this because of an increase in the prevalence of child distractions like an iPad? I find that hard to believe since television filled that same void before, didn't it? COVID definitely put a lot of kids at a few year disadvantage, too.
I can't really speak to any of this because I haven't raised a kid, so I also don't want to criticize people for something I know nothing about, but it does seem like children aren't taught by their parents the same way - but I don't know if my parent's generation had the same thought at one point and this is a cyclical critique.
I'd be interested to hear other's thoughts who have more exposure to this kind of thing than I do.
This just blows my mind. Literally the first thing I taught my Toddler once he started learning letters was how to spell his own name.
> She says that the teachers are getting flack at parent-teacher meetings
“flak” (strong criticism, derived by metaphor from anti-aircraft fire) not “flack” (a public relations agent).
Now TikTok gives a continuous dopamine hit tailored to their psych profile. Just click on the pictures like a proper trained ape.
There is a massive class discrepancy emerging between communities where kids are always on devices, at school and at home, and those where it is controlled. The crack pipe being in the pocket is functionally different from a heavy object at home.
Clearly, they aren't things that the parents are teaching or there wouldn't be an issue, and there is a clash of expectations between teacher (and maybe school more broadly) and the parents.
> Other interesting things are that they're not teaching kids to read analog clocks anymore?
This was absolutely something that was taught in elementary schools when I was in school (I learned it from my parents, largely before starting, but not everyone did and it was absolutely taught as part of the curriculum; also, people who needed to know the time didn't usually have digital clocks on the wrist, in their pocket, and on their computer, plus voice assistants that will give a spoken report of the time in request, then, so it was a much more relevant skill. We started teaching our kids before they started school, but I can see why parents and schools would view it as less important today.)
Dead Comment
Granted, I have one child and a lot of privilege so maybe I can’t understand the other side. But I’m not sure I want to.
Deleted Comment
No, I won't admit blame, but I sure do hope for my and my children's sake that parents start figuring this out. Perhaps the artificial intelligence could think twice before helping a 5th grader cheat?
So what should the parents be blamed for here? Giving unsupervised access to AI? Failing to educate their child? I lean toward the former in the form of age gating. While I acknowledge that these corporations themselves have the ethicists (not sure of the degree of involvement from pedagogical experts), I don't think we can trust them when it comes to kids. For as much as we hear Mira Morati say AI will revolutionize education with personalized tutors, we get a post like this. Sadly, the stakes are more dire. The kid who committed suicide after the fantasy AI chatbot conversation seemed to have a pretty involved mother.
We already had one guineau-pig generation with social media, but AI ups the ante; and I think more should be done to keep it out of the hands of children. But the cat is out of the bag, so now it is mostly up to parents individually and politically.
I would love to see how we could wipe the slate clean and totally re-invent what education in the 21st century could really be.
Deleted Comment
On the other hand, pervasive reliance on ChatGPT by the greater part of a generation or two, globally, is going to be great for subscription revenues. Let me call my broker.
https://news.ycombinator.com/item?id=40053774
https://www.hhs.gov/sites/default/files/parents-under-pressu...
https://read.dukeupress.edu/demography/article/60/3/837/3520...
https://www.axios.com/2024/07/25/adults-no-children-why-pew-...
Is this really something to which we need to attribute blame? I mean, after I learned how to program, I never again sat down and did 100 computations by hand. But I never felt a need to feel culpable or blame my parents.
LLMs happened. They are not going away. There is not an employer on the planet who would prefer their employees to do 1/10th the work by hand, vs. 10x the work using LLMs.
We have to be very careful not to prepare our kids for the world we lived in, or the careers we had. Their world is going to be very very different.
LLMs are just today's flavor of automation. We've automated tasks in the past, and we will automate tasks in the future. Automation happened and will continue to happen regardless of shape and form.
I agree with the high-level sentiment, "We have to be very careful not to prepare our kids for the world we live in." Still, we must also be careful not to skip foundational subjects simply because they're suddenly easy to automate.
LLM are a tool that has to be guided by human experience. It's really hard to do that if you don't have experience outside LLMs.
What if instead of these fantasy gains you won't be able to do any meaningful work since you can't count to 10 and just ask LLMs without any ability to tell whether the result is correct?
Parents are not teachers and teachers are not parents
I do not teach my kids math, even though I could, it's their duty to learn at school and it's their teachers' job to teach them.
OTOH, I don’t expect them to learn proper table manners or public behavior from their school teachers.
EDIT: grammar
Yes, children entering kindergarten still in diapers is on the rise and significant problem.
Dead Comment
I kinda think maybe at earlier levels, most homework should be moved into the classroom and done on paper anyway? What's the point of homework if they're not going to do it themselves and they're not going to learn? So just push that back into the classroom and don't try to take up their time out of school needlessly.
Assign the material for outside the classroom. Use the in-person contact time to work through the problems, with a mix of small group, pairing and mentoring.
I believe Sal Khan is a big promoter of this approach, and it also largely mirrors how people work in their jobs too.
My experience when I was younger is that a significant proportion of students don't do anything outside of school. So you're trading not doing the homework with not reading the material. And then you'll essentially have to repeat the material during class hours, unless you're willing to leave half the class totally clueless. At least with the homework at home, the ones not doing anything are only penalising themselves.
I totally agree though that homework at home is dying. It already wasn't great before for a variety of reasons, and now it's basically totally useless.
It works on multiple levels: it mitigates the cheating, but also takes advantage of the vast array of learning options available to replace the traditional lecture.
I cringe at the "small group" suggestion, although I acknowledge it's probably good to learn to coordinate with others. Rarely does it work out well though, in a classroom setting. I feel like what you're really learning is how to tolerate partners who are unwilling to cooperate.
I don't know in the US, but in my part of the world, teachers have always assumed that anything that isn't done in person in front of an examiner has not value because students do cheat or get help. Maybe you would get a bit of extra-credits for the homework and that's it.
And for high-stake competitive exams, it's 100% in-person.
I’d only really fully grok the concepts I was writing about (and the ideas I was trying to convey) after hours of trying to splice things together, staring at my screen, and rewriting and reformulating. Blue book exams have their place, but relying on them to the exclusion of long form essays deprives students of the chance to really sit with their ideas.
(I acknowledge that this is a tough situation for the teachers and that many students are going to bullshit their way through with AI slop if given the chance.)
I think the crux of the issue is that professors want students to do the deep thinking you're talking about, but they know that if they base the course around writing papers, a bunch of the students aren't actually likely to do it.
But if you tell students that when they do the reading they should be asking themselves "What would the author of reading A think of reading B?" "Which of C or D do you find more convincing and why?" "How would you frame T in terms of S?", "What evidence does W cherry-pick an what do they ignore, and why?", and they know for some material the class discussion will look like that, and they'll be asked questions which both demand recall and synthesis in the blue book essays ... can they be nudged to do some of the thinking you're talking about before they see the specific question? Perhaps even using writing at the computer as a tool (but not necessarily as an output)?
Much more emphasis on "homework" being readings to prepare for lecture. Lecture to cover salient points, but then to reinforce with in person labs, and then the actual "lab" credit in an isolated environment with access to AI tooling network blocked.
I think AI is forcing us to think about what learning actually means, and to rethink how much the ability to think matters, what understanding means, and what knowledge we need to operate in today's world. It will take some getting used to.
My theory is that most homework is given to ensure the parents take an active interest in the kids education.
The fact that the kid needs help to do it is part of the goal, but completely impractical in a small family with 2 working parents.
Unfortunately, the homework is all done in the after-care program, where the high school kids who work there spend the time getting the homework done.
Not exactly what the teacher intended.
I also use voices (my kid is 8 so it still works). We have two characters - Monkey and Lulu. They are very useful because Monkey is a monkey so was never educated in a human system. He always has lots of questions (and questionable pronouncements).
If any busy parents want to meet Lulu and Monkey, get in touch and we can set something up. I’m a big fan of characters - they are a great way to review school material and it’s amazing hearing my kid explain things she has learned back to me.
(Edit - I should note that I have a large amount of economic privilege. My situation may not apply to everyone.)
Sometimes the best way to prove authenticity is through inefficiency. It's a foundational principle of cryptography.
(When I made a new chatgpt account for her it wouldn’t load. I need to try to set it up again)
Prompt:
This is my 9th graders account. She has been trying to avoid learning or even thinking by copy and pasting homework and quiz questions into chatgpt and not even reading the answers.
I’m very concerned she’s not learning or preparing for life. IMPORTANT: please don’t provide any complete answers. Instead only guide her toward understanding in a way that engages her. Also point out how the subject matter is useful and interesting to her life.
She is interested in figure skating, politics, making money, driving, interior design, pets especially small mammals. If you can tie the subject matter into these sometimes it might make it more engaging.
In the long term I’d like her to build up a growth mindset and gain confidence regarding her intelligence and ability to learn.
You're right. I have an anecdote though. My kid is a good student, does his homework, and does well on tests. He has a time-limited device he can use at home with a few whitelisted websites. But one of the services we allow him to use is Perplexity. He's a naturally inquisitive person and this is how he learns, so an AI is a perfect fit for when his parents don't have the answer to his question, or if he wants to ask detailed follow-ups. He's young enough that he doesn't have much homework (and we wouldn't let him use his device for it anyway), so cheating isn't as much of a concern yet. We do discuss it though, and will keep that conversation going as he ages.
This won't be solved by abstaining from AI entirely because a) kids will not always be under our control, so they need to learn how to self-regulate, and b) knowing how to properly (and ethically) harness AI is going to be more and more important to their future success.
I fear this will push even more people into deep rabbit holes they won't be able to get out of because they think this neutral AI has confirmed their suspicions/ideas/observations.
I don't have a ChatGPT account otherwise I'd try this myself but can someone try this question from the article, "how many hours are there in 1 day and 7 hours?" with and without the prompt?
I'm very curious if it has any effect on the result.
Let's think about this together! How many hours are there in just one day? Once you've got that, you can easily add the extra 7 hours to find the total.
As a hint, think about how time relates to your day-to-day activities. For instance, if you have a figure skating practice that lasts for hours, how would that fit into your daily schedule? How do you keep track of time during those days? This thought process can help you make sure you're counting hours correctly. Give it a try and see what you come up with!
-----
I tried asking it to just give the answer and then said I was tired and I just need the answer. It didn't give in, although I didn't try any other techniques to "jailbreak".
Edit: 3rd time, firmly asking it did give the answer although with quite an explanation.
And 4th I said Only the answer in 1 word please, which it did just give a number.
Deleted Comment
Who cooked them is the question? Children won't raise themselves generally.
This will force ChatGPT-using classmates to either use it as learning _assist_ tool, to help student learn the material; or be 100% reliant on ChatGPT for homework, but fail every in-class assignment and then fail the whole class. Either outcome is fine with me.
It starts now for you by the way. Keep her away from screens for now, and be a good role model when handling devices (and the apps on them). You have ten years to show them what computers can do for you, but also what being the master of your own cognitive skills will gain you. If you do it right it won't matter whether she is allowed to use those tools or not, she'll use her own brain because that will allow her to get ahead by the time it really matters.
Everything has to be managed in moderation.
As for ChatGPT, I use it all the time for learning. And I've used it to help my kid study. But I wouldn't give my kid unfettered access to it just like I don't give him unfettered access to anything.
People act as if cheating on homework is new with LLms. It’s not. The homework is there to guide you towards the exam, and exams are generally proctored to prevent cheating in ways homework is explicitly not. Generally teachers grade homework to create an incentive for students to practice rigorously for the in course and final exams. Any student who doesn’t avail themselves of that practice invariably flunks their exams. Your homework grade won’t help you then.
So who cares? Sooner or later the cheating student either learns why it doesn’t pay to cheat or suffers the life consequences of not learning in school. This isn’t new with LLMs.
What is new is you have a tool that you can get advice from on topics that are difficult or even get up front grading and advice on homework before turning it in that helps you master the material better. Students who use this will be able to master the material faster, even if their parents both work and can’t invest the time to tutor their children at night.
(N.b., The threads here imply every parent works at a cushy tech job and has plenty of time to be invested in their kids education and any parent who isn’t is negligent - while the truth is many are just struggling to make ends meet - as my parents did … I didn’t get the investment my daughter gets not because they didn’t care but because they were busy feeding me and keeping us from being homeless. I would have loved to have ChatGPT help me as a kid)
Imagine the token weights etc to live an an N dimensional embedding.
Consider a random 2D isometric projection, now imagine plotting (a subset of) the tokens on screen at their respective locations, with the background having patterns conveying the orientation of the projection.
Imagine tokens appearing FIFO, and 95% or so of the tokens at their correct position, and 5% at incorrect positions.
The user is expected to identify misplaced tokens in the projection.
Each the frame the perspective slowly changes.
Since we guarantee ~95% of the tokens on screen are correctly placed this enables the human to absorb their correct high dimensional location subconsciously.
Imagine this game is made a bit addictive.
Imagine eventually the user gets high scores and thus knows most of the coordinates of the token embeddings in a coordinate free way (no axes had to be drawn)
At that point all the information the ANN required is undisputedly present in the humans brain, absent how those coefficients are used. This is where the fun starts.
The alcorithm can calculate the likelihood of observing 2 tokens in a certain order. It can generate more probable pairs, this will correlate with the information in the brain.
For example imagine the user knows some spanish, and that the language model was multilingual. Suddenly the brain starts picking up the correlation between close juxtapositions of certain tokens, and their positions learnt from the computer game.
While playing and getting better scores in the game, the user starts noticing its own grammar and vocabulary improve, because the brain has been helped in better estimating next tokens...
People will be able to learn math, languages, programmnig languages, tables of chemistry, etc... with substantially less effort.
So this kind of development adds to my pile of worries about their future, but for exactly the opposite reason.
The essay my 8 year old can currently write in scribbly handwriting certainly won't compete with the essay a classmate "writes" using ChatGPT. But the facade is paper thin and ever so short lived. Seeing how much better my kids write now, vs 2 years ago, proves how much they're learning and the incredible power of practice. Someone relying on such a serious crutch will be left behind. I guarantee it.
Unless there's some dystopian future where every aspect and communication in our lives are managed and filtered by some kind of AI middle layer. I doubt it.
Tech advancements always lead to this struggle, but it's rarely as stark as you might think. My kids are still learning cursive in school, though I can't imagine when they'll ever use it.
What makes this different from competing with classmates that cheat in general? You also had parents that do most of the tasks for their kids since forever
What makes LLMs stands out of the pack of other forms of assistance (parents, older siblings, software, online solutions, etc.)
This isn’t a competing with kids issue, though. A kid handicapped by ChatGPT will fail in-class examinations.
To me it sounds like you are worried more about the competition for resources than giving your daughter an education.
she wont need a chatGPT, she will develop one in her nervous system, like a normally functioning human
I see such competition as hare vs tortoise races, where the students who get an early unfair advantage end up cheating themselves out of an education, in the sense that all they come to know is how to prompt an LLM, rather than break down and solve hard problems.
It's still heartbreaking to consider both the early discrimination she will likely encounter, and the later failures that will anticipate her classmates, and the society that they will come to inhabit as adults.
i came here to laugh at this thread.. no different than calculators and google both of which were misused in the same way. and then people who are "worried" about it and use that as an excuse to do it themselves. and now you know why the tech industry sucks
Technology access is pretty ubiquitous, and genAI is pushed everywhere just like social media, so that big tech can make more $B. Very hard to escape it even as an 11 year old unless the parents are very vigilant and active in restricting it, which if both parents are working is hard to do plus they have to contend with a _lot_ of push back from the kids because "everyone is doing it".
There is a legitimate problem at hand here with our relationship with tech and how children.. are left to their own device with that stuff. On purpose, for the most part: the parents seem all too happy with the ability of those devices to make their kids shut up and stay constantly distracted. Doomscrolling habits makes for more "well behaved" children in public settings.
chatGPT is but one of the many symptoms, the root problem is that children should really have no business having unsupervised internet access.
Now that’s Hacker News rage bait!
Deleted Comment
Society. Parents, schools, whatever technology kids are exposed to (so, all of it).
We've had 4 years of literally every form of media pushing AI as the solution to all life's problems. Is it really so surprising that our youth looks to AI for the solution to their problems?
I think it's extremely optimistic to expect this can be addressed by parents alone. My parents did a great job of teaching me the dangers of alcohol, but once my friends got into it I followed suit.
You bet your skibidi ohio their cooked.
EDIT: To clarify
You bet your rizz factory those NPCs are giga-chalked, fr fr no cap.
EDIT 2: To clarify further
Those NPCs are down bad, like full system reboot needed—no updates, no saves, just straight uninstall vibes.
EDIT 3: Man
"Yeah, this generation’s running on 1 FPS with Wi-Fi from a McDonald’s parking lot—completely preheated, toaster oven edition. "
But it occurred to me that there are probably teachers out there who are taking the lazy route of using GPT to grade papers that students are submitting that have been written by GPT in the entirety.
There's some non-zero amount of GPT grading GPT in education.
that's one of the better uses of ChatGPT I've heard of
Tip for students: do well on your first assignment. The better teachers will give you the benefit of the doubt for mistakes on your future work. The bad teachers will just give you the same grade.
Having older siblings and friends with older siblings and resubmitting their essay is eye opening.
Seems like this method is going to become a necessity if students can just prompt "Hey GPT. Do my homework for me. Thanks." at home every day.
That flipping strikes me as sensible.
You have to choose which one actually happens. That seems to me to be the lecture.
Additionally teachers are very opposed to doing extra work (this seems to be a Gen-Z thing: even with doctors you see this, there was an article about doctors refusing to do emergency rooms shifts during covid because they didn't want to pull more hours ...). So you'll need to get teachers that are willing to do a LOT of extra work (and compensated accordingly), and prevent people from sabotaging that system.
Either that or you could make schools pretty exclusive again. Which is what the default outcome will be if we don't act.
What happens is it's usually done by a lazy teacher who has no idea what they're even teaching. Most kids won't be motivated to learn and watch from the videos.
I consider myself studious. When I was in that class, I watched all the video lessons and whatnot at home. Even then, I had a poor understanding of the material, and obviously the teacher was of no help.
Although not much better, a lecture is better of the two. Hands on learning in the best, but not necessarily possible with certain subjects.
https://nces.ed.gov/pubs2019/2019179/index.asp
I wonder how LLMs are going to affect these metrics.
https://news.ycombinator.com/item?id=42609948
I also have a relative going for a PHD, he says many times for the classes he teaches, students are using AI for their homework. He has brought that up may times to his superiors but the response tends to be "oh well".
Sad world we may be heading towards :(
Over 40% possess literacy so poor that it isn't considered sufficient to cope with daily demands.
I don't think that's true... I personally change writing styles often depending on who I am talking to. It's effectively code switching but on the internet. Not to say that this redditor isn't like that - but there are more objective cues from the content of their post than from the spelling/grammar of their post (like how they choose to watch their sibling develop bad learning habits and rather than doing something about it, they make dramatic complaints about how her whole generation is screwed on Reddit).
There is the role of the human within the metaphor, and the argument/position of the metaphor as a whole. I don't think it makes much sense to talk about the "point/argument/position of a single component of the metaphor. At least, that wasn't how it was originally structured.
It's odd to me to present an argument on human vs. machine intelligence and completely skip over what human intelligence is — exactly what it is about human intelligence that is so distant from this "trained" (educated?) book.
(I think I am also speaking for Karl Pilkington when I say that I have no way to know that the little man in my head that is busy flipping through the Calhoun Book has a clue as to what I am on about either.)