Writes Cal Newport, in "What Kind of Writer Is ChatGPT? Chatbots have been criticized as perfect plagiarism tools. The truth is more surprising" (The New Yorker).
Struggling to write the final line to his essay, Newport asked ChatGPT. It offered: "In the end, the true value of tools like ChatGPT lies not in making academic work easier, but in empowering students to engage more deeply with their ideas and express them with greater confidence."
I quoted that last line and asked ChatGPT to write a 1 or 2 sentence blog post reacting to it. I got: "This perspective highlights the transformative potential of tools like ChatGPT, not as shortcuts, but as catalysts for deeper intellectual engagement and self-expression. Embracing technology in this way can truly empower students to explore their ideas with newfound confidence and creativity."
I quoted that last line and asked ChatGPT to write a 1 or 2 sentence blog post reacting to it. I got: "This perspective highlights the transformative potential of tools like ChatGPT, not as shortcuts, but as catalysts for deeper intellectual engagement and self-expression. Embracing technology in this way can truly empower students to explore their ideas with newfound confidence and creativity."
Bleh.
39 comments:
The students don't read so they write scrappy essays. Even AI can't fix that.
Are you sure that answer was from ChatGPT and not VP Harris?
Crappy, not scrappy.
ot: On The Jack BS:
Respected DC Lawyer says:
"DJT DC case:"
“Facts” & allegations “supported” by evidence (testimony, statements, documents) that hasn’t been subject to cross examination or confrontational techniques are not properly or constitutionally established.
They thus cannot satisfy the govt’s burden of proof." 8888
&
How to write word salad that has an appearance of great writing. Makes sense, but lacks any inspiration. In other words it’s how to filibuster anything.
The Atlantic: "The Elite College Students who can't read books"
Profound but hollow.
ChatGPT sounds like an Anthropogenic Intelligence and sociopolitical activist. The value of MCACs supplemented with semantic primitives is information gathering, assembly, and presentation. A challenge is to avoid short-circuiting development of the human mind, especially at a young age. A further challenge is that reliable energy required to build MCACs is a first-order forcing of climate change or a nuclear solution that is fraught with risk in the modern political climate.
"The Boomer perspective highlighted the transformative potential of tools like pot, acid, speed and coke, not as shortcuts, but as catalysts for deeper intellectual engagement and self-expression. Many believed embracing pharmaceuticals in this way truly empowered students to explore their ideas with newfound confidence and creativity."
You do know, of course, that ChatGPT grew up in a middle class family.
Right?
It sounds JUST LIKE Kamala Harris spewing her word salads.
That reeks of heaviosity!
People are going to have to get over it. Writers aren't writing with quill pens on their mistress's ass any longer, or even using typewriters, or word processors alone. Sure there are going to be hacks putting out prose that reads like processed cheese, but in the hands of a person who cares about the product that he is putting out, it's a revolutionary tool.
I asked ChatGPT how Ann Althouse the bogger, would write on the relationship between Catherine of Braganza and the Duchess of Cleveland. (Catherine of Braganza was the wife of King Charles II and the Duchess of Cleveland was his chief mistress i the same period of time.)
" ChatGPT said:
Ann Althouse, known for her incisive and often witty commentary, might approach the relationship between Catherine of Braganza and the Duchess of Cleveland with a blend of historical context and contemporary relevance. Here’s a speculative description in her style:
Catherine of Braganza, the Portuguese princess who became Queen of England through her marriage to Charles II, navigated the turbulent waters of 17th-century royal politics with grace and tenacity. Her influence stretched far beyond the court, intertwining with the lives of those around her, including the Duchess of Cleveland, a figure emblematic of the complex social dynamics of the time.
The Duchess, born Barbara Villiers, was one of Charles II’s many mistresses and wielded considerable power and charm in her own right. The relationship between Catherine and the Duchess was marked by rivalry and an unspoken understanding, both women vying for the affections and attention of the king. In this delicate dance of love and loyalty, we see not just the personal struggles of two women but also a microcosm of the broader societal norms and expectations placed on women in power.
Althouse might delve into how this relationship reflects the complexities of female agency within a patriarchal system, exploring the nuanced ways these women exerted influence and navigated their positions. With her characteristic sharpness, she could highlight the irony of their coexistence—Catherine, the legitimate queen, and the Duchess, the favored mistress—each representing different facets of femininity and power in a male-dominated world.
In a modern context, this relationship invites reflection on how women’s roles have evolved yet remain entangled in similar struggles for recognition and respect. What does it say about our current societal dynamics when we look back at these historical figures, and how can we draw lessons from their lives?
This blend of historical insight and contemporary analysis would likely resonate with Althouse's readership, inviting them to reflect on the enduring themes of power, gender, and rivalry.
I found this ChatGPT attempt to write in the Althouse style very interesting. Would Althouse "delve into how this relationship [between a Queen and the mistress of her husband, the King] reflects the complexities of female agency within a patriarchal system, exploring the nuanced ways these women exerted influence and navigated their positions." Or would she refuse to pay any attention to King Charles because he wore shorts?
https://www.factinate.com/wp-content/uploads/2019/02/charlessquarethumb4.jpg
Would RH Hardin reflect on "enduring themes of power and gender", noting that the Duchess of Cleveland had political power even when women did not have the vote?
Would sociological jargon of the kind AI formulates help me understand early modern England ? What would Andrew Marvell, "standing between the morning and the flowers," say to Chat GPT?
ChatGPT is a sociopolitical animal with religious entrails laid by anthropogenic intelligence. Gender refers to sex-correlated attributes: masculine and feminine. What formula would it conjure to be equitable and inclusive of the transgender spectrum? Is it albinophobic?
When I’m crafting a reply and predictive aI correctly guesses most of my next words it makes me wonder if I’m actually doing the writing.
"Embracing technology in this way can truly empower students to explore their ideas with newfound confidence and creativity."
The use of the word "truly" is almost always a sign of bogosity.
Lots of excitement about The Great Replacement: A-YiYi.
I feel as though ChatGPT is pulling my leg in suggesting thatistheAlthouse style
I feel as though ChatGPT is pulling my leg in suggesting thatistheAlthouse style
There is something so earnest about most llms responses that makes them stand out. Even if you tell them to be sarcastic or cynical or worldy they still sound super earnest. Like the brightest student doing an assignment to make the teacher happy. Maybe it's the lack of self-awareness.
There have been a bunch of these "ChatGPT is really not that bad" stories popping up recently. I assume a large fraction of them simply paid for by tech companies, but not all, and probably not this one.
I think the reason is that we are just at the point in the semester where the first student attempts at papers have come in and been graded, and teachers of writing are now seeing just how badly the covid years degraded the writing ability of the students currently in college.
I just finished grading the worst set of papers I've seen in 28 years (actually, in 32 years, since I didn't see papers this bad when I was a graduate T.A. teaching English 101 in 1992. These students are SMART, motivated, full of curiosity and a desire to learn. I LOVE teaching them. But they have many writing problems, and many of those problems are weird things we haven't seen before. I've gotten digressive and run-on paragraphs thousands of times, but never paragraphs in which the sentences were just thrown in at random, which I saw in multiple papers last week. I've never seen so many basic words misused so badly ("prolific" to mean "important"; "valid" to mean "acceptable to some people" ; "arguably" to mean "incorrectly"), or so many failures of basic logical reasoning. These students aren't spewing out politically correct or work gibberish (at least mine aren't), and they're not holding on to over-personalization like students from, say, 2015-2019 did. But they clearly have not received enough honest and detailed feedback on their writing and seem in any ways to be flying blind.
And I think there are started to be more and more professors who think "I can't fix all this weird crap--but ChatGPT can." They would like the fig leaf of "experts say" in order to offload teaching students these really basic writing skills that can only developed with attention, frequent feedback, multiple revisions, and a massive increase in the range of materials that students are reading. So there is now a market for "research" (but really just opinion) that will "demonstrate" that you aren't doing your students any harm by actually encouraging them to use LLMs for their writing--and it'll make your grading easier, as you won't have to give any time-consuming feedback on individual sentences or paragraphs.
And then professors, having taken yet another step to reduce the time and effort they spend on their students, will wonder why they continue to lose compensation and respect.
My reaction exactly.
Both.
Perhaps everyone but me has missed the elephant. Cal Newport is another but even more banal AI chatbot.
Welcome to the age of recursive mediocrity.
Having read a few 'greats' (which demonstrated my intellect is not capable of enjoying such works) there are similarities, but you can clearly see an emotion, skill, and storytelling in human written works.
Dumbing down the education has a specific goal of making AI written works indistinguishable from the real. This supports an appeal to authority that democrats so love.
I missed this one yesterday. I recently subscribed to ChatGPT because I had reached my daily usage limit in the free version. I’m going all in on AI. The dismissive tone in this comment section seems pretty misplaced to me.
Professor Drout: I was amazed in the mid-1990s to discover that many of my freshman students at a competitive university had never read a whole book, and among those who did, it was Stephen King or Michael Crighton. Most could not write competently. Then again, for decades, the university had been treating reading and writing skills with similar neglect by using virtually untrained graduate students to teach all the composition classes (for pennies) so that the tenure-track professors could lounge around on their hobbyhorses and occasionally publish incomprehensible and unread essays and books in the argot of ludicrous literary theory or identity politics.
A cord was broken with the past.
ChatGPT is just the latest iteration of subsequent mass ignorance. I like the way you describe the specific effects on your students. You must have read a book or two over the years. Soon we can do away with both students and teachers, and the machines may analyze Macbeth among themselves.
My first degrees were in English lit and Russian lit. I’m one of the most well read SOBs in the classic lit you’ll ever meet, and I’m all in on ChatGPT.
GPT and AI are nice conveniences when you feel like it, as a novelty and variety. My problem is many instances you can't opt-out. It all feels very compulsory.
Calculators deprecated counting. Computers suspended auditing. MCACs with Viterbi decoders and semantic primitives outsources thinking. Progress.
You had to be there.
I think I'll stick with my native wit. Such as it is.
Shouting: you read the books; you read the classics. You are educated already. Play away: you have the context to enjoy this tool, just as advanced mathematicians enjoy exercising their hard-acquired skills.
Now picture relatively smart young people who have never read books, not even in high school. In my teaching experience, I found that the pedagogical practices of substituting snippets of books, technological crutches, and loads of chattering about so-called critical thinking over solid, foundational, historically situated reading leads to sadly presentist, bored, and narcissistic students.
I'm not alone in thinking this. And it is not just humanities classes that are being affected. Legal writing, for example, has suffered the same loss of quality, depth of reasoning, and historical fluency.
It's the wine connoisseur vs the adolescent alcoholic. Maybe we need a "thinking age" like the "drinking age" and nobody under that age should be allowed to use computers.
I agree with ST, though. I am also all in on AI, I also have read many books "sraight through!" Only a nitwit would use it to write for them, though.
"Collaborating with A.I. can also offer you a high-tech 'shitty first draft,' allowing you to spend more time editing bad text and less time trying to craft good text from scratch..."
Paint by Numbers for writers.
ChatGPT is today probably a better writer than 80% of freshman college students, maybe only 50% of the top 100 colleges.
K-3 for learning to read, and 4-6 for learning to read books. 7-12 to read books with ideas that should be remembered. Sometimes even books when they’ve already seen the movie—tho knowing so much from the start lessens the experience & excitement. Like Harry Potter books.
Early ed failure can’t be cured by college without giving up more thinking about
Ideas.
Post a Comment