It's the middle of the summer. I'm feeling lazy. What do lazy people do in this century and decade?

Why, they cheat, of course, and use ChatGPT. If the university students can do it, so can I.

Besides, testing AI is part of our brief around here. Even if some of our comments cause Robbie to say, 'But, but, but. . . '

To start with, I asked Copilot if it had heard of this column. It had. (It has internet access.) I asked it for a review and got this. I swear, hand-on-heart, I would have shared it if it had been scathing. That would have been much funnier than what I got – which worries me.

The weekly column by Dmitri Gheorgheni in the h2g2 Post, titled 'Deep Thought,' is a thought-provoking series that delves into various topics with a blend of personal insight, historical context, and cultural commentary. In a recent edition, Gheorgheni explored the collapse of the Francis Scott Key Bridge in Baltimore harbor, using the event as a springboard to discuss broader issues of infrastructure, engineering, and societal responsibility.


Gheorgheni's writing style is characterized by its depth and accessibility, often starting with a specific incident or observation and expanding into a wider philosophical or social exploration. The column's name, 'Deep Thought,' reflects the contemplative nature of the pieces, inviting readers to ponder alongside the author on the complexities of modern life.


In his columns, Gheorgheni combines storytelling with analysis, drawing connections between past and present, and often referencing literature, history, and science to enrich the narrative. The result is a column that not only informs but also challenges readers to think critically about the world around them.

It concluded with a 'learn more' section with relevant links, including one to this column's archive. Colour me impressed. I mean, this is downright scary.

No, not the fact that a robot said nice things about me. The fact that the prose generated made sense: it was coherent and, above all, relevant to the subject at hand. It used examples from text.

And yes, its research was less in-depth than it let on: it picked one example and extrapolated from there. (Hm, how did it pick that one?) It seems unaware of the connection between the words 'Deep Thought' and any literary source which might have served as possible inspiration for the title. Nonetheless, the result of that idle question was a disturbingly good answer.

I'm with the university faculties of the world, who are alarmed at this development and worried that their students will turn in ChatGPT-generated papers rather than doing their own work. If the result of their attempts is this good, lotsa luck, profs. I mean, it's not Shakespeare, but it's good enough for freshman English.

There may be clues, though. Copilot's mini-essay on 'Dmitri Gheorgheni as Columnist' was the result of a careful prompt. I wrote, 'Please describe the weekly column by Dmitri Gheorgheni in the h2g2 Post called Deep Thought.' I pointed it in the right direction. You think that would be simple. You don't know lazy undergraduates.

I once had a semester paper to read that informed me that the Romans liked to watch TV in the evenings. This unobjectionable statement came from a guidebook to modern Rome. The problem? The paper was supposed to be about ancient Rome. My mind boggled, as they say.

Another student turned in what was supposed to be an eight-page paper. It was three whole pages long but only by dint of triple-spacing. This brief masterpiece concerned Cleopatra – rendered throughout as 'Cleopatria'.

If students could accomplish these things in the age of word processors and floppy disks, what new wonders will unfold in this brave new world of 'your AI pal who's fun to be with (and incidentally is contributing to global warming)'?

I'm sure the professors will find a way to cope. Education is an ecosystem: as student deviousness evolves, so does educator response. It's a predator/prey situation. I refuse to speculate as to which is which.

Now for another test: can it generate one of my columns? Let's see what it does.

Prompt: Please generate a two-paragraph essay in the style of Dmitri Gheorgheni about AI.

Result:

In the labyrinthine corridors of the digital age, where the binary beats of ones and zeroes compose the symphony of our era, artificial intelligence (AI) emerges as the maestro, orchestrating a world where complexity is rendered into simplicity. Dmitri Gheorgheni might muse upon this phenomenon with a blend of skepticism and wonder, pondering the implications of these silicon sentinels in our midst. AI, he would argue, is the mirror reflecting our own consciousness, a simulacrum of sapience that challenges us to redefine the boundaries of life and intelligence. It is a creation that dances on the knife-edge of innovation and caution, a testament to human ingenuity and a harbinger of an uncertain future.

Yet, Gheorgheni's contemplation would not be mired in trepidation alone. He would recognize the poetry in the algorithms, the artistry in the code. AI, in his view, is akin to a grand experiment in the alchemy of thought, transmuting data into decisions, information into insight. With a storyteller's grace, he would weave tales of AI's potential to elevate our species, to tackle the grand challenges of our time with a precision and efficiency unattainable by mere mortals. And in this reflection, Gheorgheni would find a deep thought indeed: that in our quest to create machines that think, we are, in fact, embarking on a journey of self-discovery, learning more about what it means to be human in the process.

I am gob-smacked. That's not half-bad. 'Silicon sentinels' is disturbingly prescient: I've been thinking about 2001: A Space Odyssey and its relation to Arthur C Clarke's short story, 'The Sentinel', which Clarke said had very little to do with Kubrick's film. I was thinking this because to me, one reason for creating AI is to have a system sophisticated enough to manage deep-space travel, far into the human future. I'm still reading A City on Mars by Kelly and Zach Weinersmith, and believe me, it's an eye-opening read about the challenges to space colonisation.

Watching AI doing an impression of you is also eye-opening: am I really the Joseph Conrad of the internet? 'Alchemy of thought,' indeed. Death's too good for it. It's right about one thing, though: we don't study AI for its sake. We study it for our own.

Dmitri Gheorgheni

15.07.24 Front Page

Back Issue Page