AI writing is here, and it’s worrying. Can writers and scholars adapt?

In recent years, artificial intelligence (AI) has made incredible progress in its ability to generate human-like text. As a result, AI copywriting is becoming more mainstream, with businesses and organizations using it to create everything from marketing copy to financial reports.

Although AI writing is still in its infancy and far from perfect, it is clear that it poses a threat to the livelihoods of professional writers. After all, if a machine can produce text indistinguishable from that of a human writer, why would anyone need to hire a real person to do the job?

It’s not just low-skilled jobs like content writing that are at risk of being automated by AI. Even high-skilled jobs like journalism and writing novels could eventually be replaced by machines. In fact, a Japanese company has already developed an AI system that can write novels better than humans.

Of course, it will take some time before AI writing becomes good enough to completely replace human writers in all genres and formats. But as technology continues to improve, the day when machines can do our jobs better is fast approaching.

The end of human writers?

The above four paragraphs were generated by OpenAI’s deep learning AI writing model, called Generative Pre-trained Transformer 3 (GPT-3).

Indistinguishable from the words of a human writer, the program can respond to any prompt entered by a user, and among many other forms of writing, it can build a short story, hold a conversation, or write a news article. .

This begs the question: is this the end for human writers?

According to Professor Mike Sharples, who has decades of experience in writing and AI research, the answer is “not yet”.

“You can either take some sort of apocalyptic view, AI is going to put professional writers out of work, everything is catastrophic and AI is going to take over,” he told Euronews Next.

“Or you can take the glass-half-full approach, which is there are amazing tools coming in and as writers we can put them to good use and as teachers , we can put it to good use”.

Sharples, professor of educational technology at the Open University in the UK, said written extensively about AI writing and its development, and he considers attempts to resist it – whether in professional writing, business or academia – to be futile.

Luckily for journalists or others who write about hot topics, the way AI systems are trained means they aren’t fully aware of the latest developments going on in the world.

Even so, while the writers needn’t panic just yet, he says they “should be worried.”

AI writing is increasingly being used to produce content on the web, especially for marketing or blog posts, as companies compete for SEO supremacy.

Freelance job sites like UpWork are seeing more and more job postings looking for writers to specifically use AI writing tools like Jasper to generate content faster.

“You can either see this as a huge boon, a huge tool to help you write faster to get your words across, to get published. Or you might see it as a threat because anyone else might,” Sharples said.

Standard plagiarism detection methods don’t work

Social media sites like Reddit are full of users telling their stories of using AI writing tools to get good grades in school or college, or asking for advice on the best tools to use. use to avoid detection.

One user, who said he was studying biochemistry in college, told Vice’s Motherboard, “For biology, we were learning about biotech and writing five good and bad things about biotech. I would send a prompt to the AI ​​like, ‘what are the five good and bad things about biotech?’ and that would generate an answer that would give me an A”.

A job that would have taken them two hours now took only 20 minutes.

The problem for those who set and assess academic writing tasks is that even the newest plagiarism checkers can’t keep up. According to Sharples, attempting to do so would be “a futile computer arms race”.

“AI doesn’t just copy things from the web, it actually creates new text,” he says. “It’s inventing new ways of expressing yourself. The standard methods will therefore not detect it”.

He is not infallible. Sharples used AI to write essays that, on the face of it, seemed entirely plausible and would go undetected by a plagiarism check. However, there were a few flaws.

The AI ​​knew to include references like any good college essay, but on closer inspection it turned out that some of the references were made up. Other quotes included by the AI ​​were actually taken from studies that had supported the exact opposite of the argument made by the AI.

The other, more obvious way for assessors to detect if students are using AI writing tools, he added, is if the quality of their writing suddenly improved.

Instead of trying to fight it, Sharples thinks that just as professional writers need to accept that AI writing is here to stay, educational institutions need to do the same.

He argues that educators and policymakers need to rethink how to assess students, and that AI systems could help students learn to become better writers.

They can be used, for example, to quickly show students different ways to express an idea, or as a creativity exercise, where students can write a story in tandem with an AI tool.

How does AI writing work?

Sharples describes the AI ​​writing model used by OpenAI as a “highly bloated text-completer”.

Just like your phone does if predictive text is enabled, AI writing models look at what’s been written before and predict what’s next. But whereas a phone’s predictive text examines the last characters typed, OpenAI’s model can look back at the last 700 or so words and generate hundreds of words of suggested text.

And what he writes makes sense, because he was trained on almost all the written texts available on the Internet. He knows the context in which he is writing and is therefore generally indistinguishable from text written by a human,

“He was trained on Wikipedia. He was trained on blog posts. He was educated on e-books, online books and world literature. So it uses this vast database,” Sharples said, explaining that it creates an internal “mental memory” of how language is processed.

And it’s not about mindlessly regurgitating text either – it can come up with new ideas.

“That’s what even the developers of these systems still don’t understand – that it’s not just about repeating text,” he said.

“It’s not just about taking previous words and reusing them, but creating an internal representation, not just on the surface of the text, but of the ideas and concepts behind it.”

“He creates this neural network, this multi-layered network. And we know that some of those layers are about the words, the style, but other layers are about the way the text is structured and the content, the underlying ideas.”

All of this has major implications for more complex and time-consuming writing endeavors, like writing books.

A recent article on the OpenAI community forum details one member’s use of GPT-3 to write an entire 38,000-word book on proverbs, platitudes, and truisms.

“I used GPT-3 to generate many lists of proverbs and quotes from around the world, then reused it to write a brief description of each,” user daveshapautomator wrote.

“This book contains more than 600 proverbs and quotes. Sent it to a friend for a beta review and will format it for printing while working on proofreading with Grammarly. All in all, it should only take a few weeks to go from first draft to print.

The future of AI writing

The flourishing of AI authoring tools has occurred alongside the release of a number of other AI authoring tools.

Facebook parent company Meta recently unveiled an artificial intelligence tool that creates GIF-like videos from text prompts.

Another of OpenAI’s tools available to everyone, DALL-E 2, creates still images from text prompts.

Sharples, the professor, sees the future of AI content creation becoming multimedia, with more sophisticated options for creating text, images and video as a package.

He also expects further refinement of pure AI writing, with AI systems “pre-training” including newer content, making them capable of writing on more timely topics. .

Plus, he thinks they’ll start blending into real-world working models, so the writing becomes even “more consistent and plausible.”

And soon, he predicts, we’ll start seeing sophisticated AI writing assistants in our everyday word processors.

He cites Microsoft as an investor in OpenAI: “Microsoft will therefore want to recoup its investment by integrating them into Microsoft Word and other tools. So you’re just going to see them used routinely, and it’s just going to be part of the writer’s repertoire.”

Leave a Reply