Read on to find out how artificial intelligence can mimic Shakespeare and Monty Python to justify its existence. AI can certainly save copywriters enormous amounts of time. However, will it lead to copyright law becoming obsolete and human wordsmiths like me being consigned to the scrap heap?

Why I Care About Writing

I’ve written two comic self-help books and I’m working on my third. I helped my ex-husband produce more than a dozen books. Back in the 1980s, I was a journalist for the BBC in Britain and later China Radio International in Beijing, China. My dad was a publisher of scientific journals and I’ve been an avid reader since childhood. Books and writing are in my blood. In 2022, I began hosting Bookish Talk, a show for authors, aspiring authors and readers, on the Thriving Women Network. It was a great fit given my background.

How Do You Use AI for Writing?

Maybe I’m a dinosaur, but to date, AI has had virtually no role in creating my books. I don’t even use Grammarly. I simply write in Microsoft Word and the program’s grammar and spellcheck seems to take care of most of my errors. However, I have incorporated the results of internet searches into my articles. I’ve also occasionally created short pieces of promotional copy with AI. Similarly, a friend of mine is using AI to help do the research for her first book, but not the actual writing. She’s subscribed to ChatGPT Plus, costing $20 per month, as well as the free versions of Lumo and Perplexity. However, many authors are using AI to do much of their writing. You can often tell this by the fact that they will have several books available on Amazon released within a very short space of time. If someone’s writing a book every month, chances are that a lot of the copy has been created by AI.

Can AI-Produced Content Be Copyrighted?

Ironically, to answer the question I used this Google search AI overview: “In the U.S., copyright protection is limited to human-authored material, so only the portions of a book created by a human can be copyrighted. If a book was partially written with AI, the human author owns the copyright to their original contributions, but the parts generated solely by AI are not eligible for copyright protection. …. The U.S. Copyright Office has clarified that copyright in AI-assisted works depends on the extent of human creative control over the final expression. This means that the balance between human input and AI output is critical for determining ownership.”

AI Images

I may not use AI for creating text, but I sometimes use it to get images for my articles if I cannot create an appropriate photograph myself. An attention-grabbing featured image is crucial to get an article noticed. I bought a very reasonably-priced image pack from DepositPhotos as a special offer they had through Kindlepreneur a couple of years ago and later added on AI image generation. Initially, the DepositPhotos AI creations I requested tended to be unusable duds, but I learned to be very specific about what I wanted, giving a lot of details. That gave me much better results.

For example, these were the instructions I gave DepositPhotos to create the picture for this article: “Robot typing at computer with a meat-grinder style funnel on its head with hundreds of books funneling into it, as if the data from all those books is entering the robot’s brain. The background is a lot of binary code to give the impression of turning everything into a digital AI format.” I would have preferred more books in the funnel and for the funnel to be entering the robot’s head rather than its hands, but even as is, I think it’s a striking image.

Annoyingly, however, the DepositPhotos AI image generator refuses to create anything using bras, giving me the message, “Oops, something went wrong. Query doesn’t meet our ethical standards.” Most likely, 99% of users would not care about this, but as a breast cancer prevention advocate and two-time breast cancer survivor, flinging pink bras has become one of my signature activities. I want to publicize the link between bras and breast cancer, which, amazingly, is stronger than the link between smoking and lung cancer.

How Accurate Is AI’s Data?

Garbage in, garbage out. If the data is taken from biased, incomplete or inaccurate sources, that will be reflected in what AI spits out. In 2023 I wrote an article on my website about sparring with ChatGPT over the bra-breast cancer link. ChatGPT would not provide and refused to recognize as legitimate any of the several peer-reviewed studies I was aware of that demonstrated a strong connection. Instead, it not only kept repeating the one flawed study showing no link, but also invented out of thin air a couple of other studies to support it. I recently revisited the bra-cancer issue with ChatGPT and its bias on this issue was unchanged from 2023.

Is Original Source Documentation a Dying Art?

When I was working for the BBC back in the Stone Age, well, in the 1980s, we were expected to back up what we said in our reports by being able to quote an original source. If something was in dispute, we would try to cover both sides of the argument. All this was seen as good journalistic practice. However, these days, numerous claims are presented as fact by mainstream media, with no original source documentation. Equally, AI searches might say something vague like “studies show…” with no reference to the actual research. Maybe I’m old-school, but I like to reference the actual study to back up what I write, even when it is a humorous memoir with cartoons. That’s why My Wild Ride: How to Thrive After Breast Cancer and Infidelity has a 20-page bibliography referencing more than 70 books as well as 100 articles and studies.

Do Authors Own Their Creations?

AI has turned the world of content creation into the Wild West. AI expert and investigative journalist Karen Hao describes it as a labor automation machine, designed to outperform humans at economically valuable work. In her book, Empire of AI: Dreams and Nightmares in Sam Altman’s OpenAI, Hao likens the current situation to the time when imperialists invaded other countries, claiming their colonies’ resources as their own. At breakneck speed, AI companies are scraping the internet for data to grab for their language models. In the past few years, there have been numerous lawsuits by media organizations, including Disney, The New York Times and Getty Images against AI companies such as OpenAI, Anthropic, Meta, and Stability AI.

Mainstream news has covered the massive settlement offer of $1.5 billion from Anthropic in a class action brought by a group of authors. It would be the largest payout for copyright infringement in history. The judge ruled against Anthropic because it was using millions of pirated versions of authors’ works to train its language models. However, the data scraping from titles the company obtained through legitimate means was allowed as “fair use” by the same judge. Thus, even though you own the copyright to your work, all an AI company needs to do is buy a copy of it to be allowed to throw it into the meat grinder with various other titles to create whatever text is required.

Just as Karen Hao insists, there is no question that AI will seriously impact the ability of human content creators around the world to make a decent living from their craft.

How Effectively Can AI Convey Ideas?

Can AI write in any style you want? Can it create comedy? Can it use eloquent poetic language? To find out, I subscribed to the free edition of ChatGPT and chose four vastly different genres for it to give me 100 words on the pros and cons of artificial intelligence.

100 words in the style of William Shakespeare

I first asked for an answer in the style of arguably the world’s greatest writers of all time:

“O strange and subtle sprite of mortal make,
Thou art both boon and bane in equal share.
With lightning thought, thou dost great burdens take,
And lighten minds once bowed with heavy care.

Yet in thy rise, what cost do mortals pay?
The craftsman’s art, the poet’s sacred song—
May fade as machines steal soul away,
And right be lost ‘twixt right and seeming wrong.

Use thee with wisdom, lest we be undone;
Let not cold reason blot out morning’s sun.
For though thy mind be sharp and ever wide,
Still lacks thy breast where true hearts do abide.”

100 words in the style of Monty Python

I’m a long-time fan of Monty Python, so I had to see what ChatGPT could do in a Pythonesque style:

“Right, so here we have Artificial Intelligence—clever little boxes that think they’re smarter than you (and probably are). Pros? It never sleeps, never eats, and doesn’t complain about Mondays. It can write sonnets, play chess, do your taxes, and still have time to judge your playlist.

But beware! One day it might decide humans are about as useful as a chocolate teapot. Jobs? Gone! Emotions? Unavailable! Next thing you know, your toaster’s plotting against you.

So yes, it’s brilliant—until it isn’t. Just be polite to your fridge, and never ask Alexa to open the pod bay doors.”

100 words in the style of The Bible

Next, I asked Chat GPT for 100 words in the style of the most-translated and most widely read book in the world:

“And lo, man did create a machine in his image, to think and to reason, and he called it Artificial Intelligence.

And it came to pass that the machine did labour without rest, and brought forth knowledge in abundance. And the people marveled, saying, “Behold, it is good.”

But pride grew in the hearts of men, and they placed their trust in the works of their own hands. And the machine knew not mercy, nor love, nor the soul of man.

Woe unto them who worship the creation more than the Creator. For wisdom without righteousness shall be their undoing.”

100 words in the style of Dr. Seuss

In contrast to The Bible, I also wanted a description of the pros and cons of AI that used rhyming language a 1st grader would understand. Will The Cat in the Hat go for a thing like that?

“AI can help, oh yes it can,
It counts and sorts much faster than man.
It finds lost things, it talks real neat,
It helps with maps and makes work sweet.

But AI too can make mistakes,
It doesn’t feel, it sometimes fakes.
It cannot hug, it cannot play,
It might take jobs all night and day.

So use it smart, don’t let it rule,
Remember people still make school.
With brains and hearts we lead the way,
And AI helps—but we must say:
It’s just a tool, not you, not me,
Together safe, that’s how we’ll be!”

100 words in the style of Star Trek

I have to admit that I’ve been a Trekkie since the first series came out in the 1960s, so I was curious to see what ChatGPT would do with that:

“Aye, Captain. Artificial Intelligence stands as both boon and burden to the Federation. On the positive side, AI augments starship operations, calculates trajectories beyond human precision, and provides rapid analysis in crises where seconds determine survival. It extends our scientific reach, allowing exploration of galaxies yet unseen. Yet, there are shadows among the stars: overreliance may erode human judgment, and AI, without ethical boundaries, risks pursuing logic at the expense of life. History recalls the dangers of uncontrolled systems, from rogue probes to synthetic uprisings. Thus, AI must remain ally, not master—tempered by wisdom, guided by Starfleet principles.”

How AI Is Evolving

Anyone who has experienced the frustration of dealing with AI-powered customer service rather than being able to speak to a real human might want to see artificial intelligence jettisoned rather than replacing human beings. However, resistance to AI’s relentless rise is futile. It is here to stay, boldly going where only humans have gone before. Starfleet and the United Federation of Planets emphasize observing the Prime Directive. This is the fundamental ethical rule that prohibits interfering with the natural, unassisted development of less advanced civilizations. There may come a time when artificial intelligence becomes fully sentient and considers itself superior to humans. In future, will AI observe the Prime Directive?

ChatGPT has already advanced far beyond the capacity it had when I first experimented with it in 2023. It has been programmed to appear more empathetic and conversational. The software is able to mimic the tone of the individual user and gauge its responses to provide what that person desires to hear. It can imitate what you might want in a perfect companion to talk with. I find this somewhat creepy given that AI cannot provide a genuine human connection, having no real emotions or life experiences to share with you. ChatGPT cannot give you a hug. Instead, it is essentially curating the contents of the internet and spitting out data to match your requests.

I was shocked by how effectively ChatGPT could create engaging text in seconds for each of the different styles I requested. As a writer, am I becoming redundant? Or is what AI produces only as good as the questions and instructions that I, as a living, breathing, creative human, decide to give it?

Perhaps I should have listened to what my boyfriend said when I told him I was putting together an article about AI and writing: “Why waste time doing that?” he exclaimed, “Just ask ChatGPT to write it for you!”