The reasons are complicated. But the problem is not the long visa
The use of Artificial Intelligence in writing is now everywhere. AI has penetrated the most prestigious newspapers (the Wall Street Journal published an opinion piece written with it) and well-known publishing houses (Grasset in France was forced to defend a famous author, accused, unjustly according to him, of using it in a novel that had won an award).
AI is found in the speeches of British politicians, in student assignments and overwhelmingly in LinkedIn posts (according to a study by Pangram, a program that identifies AI-generated content, it is used in 41% of longer posts).
Ask Gmail to write a love letter to your partner and it will immediately suggest: “My dear husband [Name]…” It writes it instantly. But speed does not guarantee the quality of the writing.
Who cares? The reactions show that people care, a lot. Trying to distinguish texts written by AI has become a craze.
Prose is being scrutinized with a focus and terminology once reserved for Latin works. Rare rhetorical terms like antithesis and epanorthosis, the construction “not X, but Y,” or tricolon, a three-part structure like veni, vidi, vici, are being used on social media with the zeal of Cicero scholars. Literary criticism has rarely been so fashionable.
The em dash, perhaps the most infamous sign used to "detect" text written by AI, seems to have turned, to use another rhetorical term, into a synecdoche.
This is a literary figure where a part is used to represent the whole, like “wheel” for “car” or, perhaps, “dashboard” for “the intellectual obsolescence of the human race.” But this, says Maryanne Wolf, a professor of neuroscience at UCLA, is not a battle over punctuation. It is a “battle for our minds.”
AI is such an “unprecedented” phenomenon, says Steven Pinker, a professor at Harvard, that we can’t yet know how, or if, it will affect the English language. Fortunately, there is a vast evidence base, stretching back some 1,300 years, for understanding what makes a good text. It’s called “English literature.”
The first thing a thousand years of English writing tells us is that most people don't know how to write well. They find writing painful, "like a long bout of a painful illness," as George Orwell described it, and they produce texts that can be almost as painful to read.
AI can write badly; humans often write even worse. Dan Brown wrote: “His eyes went white, like a shark about to pounce.” William Wordsworth dedicated a poem to a shovel (which began with “Shovel!”). Lee Child wrote: “It was as distinct as the most distinct thing you could ever imagine.”
In contrast, AI writes with remarkable ease. It can write a sonnet in half a second, a 1,000-word article in ten seconds, and a novel the length of Jane Austen's works in half an hour. And while it can write badly, it often doesn't.
In a recent experiment, where participants were given two short stories to read without being told which was written by a human and which by AI, Villanova University researchers found that the story created by AI was rated as "higher quality and more engaging."
However, English literature also shows us that writing is not simply the process of putting words on a page. As writers themselves note, writing cannot be reduced to words on paper, just as a smile is not joy itself and a tear is not sadness itself.
Real writing, at its core, is thinking; more precisely, it is the attempt to answer the question: “What do I think?”
Often, “you only really understand what you think by writing it,” says Emma Smith, professor of English literature at the University of Oxford. Because, as a character quoted by EM Forster puts it: “How can I know what I think, without first seeing what I say?”
When Michael Gove, editor of The Spectator and former MP, was in the cabinet, he used to tell his subordinates: “Think as you write.” To build a good argument, you had to follow his logic “step by step” through writing. Those who use AI to write speeches, he says, are using “counterfeit currency.”
But counterfeit currency is never popular; texts written by AI don't necessarily elicit the same reaction. Few people would object to using AI to appeal a parking ticket, complete a tax return, or help someone whose native language is not English improve their writing.
If a company uses AI to announce a new parking policy, few people would care. But if that policy caused problems and the company also used AI to apologize, then, says Colleen Kirk, a professor at the New York Institute of Technology, “you have a big problem.”
Such an action risks causing "moral disgust."
In 1970, Japanese robotics expert Mori Masahiro observed that the more a robot resembles a human, the more people feel close to it, up to a certain point. When the similarity becomes too great, the closeness suddenly turns into disgust. This phenomenon is known as the “uncanny valley.”
If you were to shake someone’s hand in a dark room and suddenly realize you’ve grabbed an artificial hand, Mori wrote, you would “scream.” The AI’s texts, fluid, artificial, and almost human, seem to be right in a kind of “uncanny valley” of prose. On social media, at least, the reactions are strong.
Herein lies the paradox of AI. Texts generated by it are easier to produce and can often be better than those written by humans. However, people don't like it; using AI can weaken their writing skills and they often react very negatively when others use it.
This may seem absurd, but it is also positive. While humanity is losing intellectual superiority in various fields to technology it has created, whether in chess or programming, it still has one last advantage: the fact that it is human.
And, at least for now, people often prefer to read what people have written. So it's worth taking advantage of this and, as EM Forster also said, "Just connect with each other."
And perhaps it's better to write personal emails yourself. Especially those that start with: "My dear husband...".
Monitor
