Generative AI and linguistic diversity in academic writing and publishing: Perspectives from World Englishes

The Great Language Mixer: Why AI Might Be Flattening Our World

Imagine the internet as a giant, bustling library where people from every corner of the globe come to share their stories. For a long time, this library had a very strict rule: to be heard, you had to speak in a very specific, polished version of English, the kind you might hear on the news in New York or London. If you spoke with a different accent, used local slang, or mixed in words from your own language, your story might get ignored or marked as "wrong." This is the world of academic publishing, where scientists and scholars write their big ideas.

Now, imagine a super-smart robot librarian has arrived. This robot, powered by Generative AI (like the chatbots you might use to write essays), is incredibly fast at fixing grammar and organizing thoughts. It's called Generative AI because it can create new text from scratch. But here's the catch: this robot was trained mostly on books written by people who already spoke that "perfect" New York or London English. It doesn't really know how to handle the colorful, messy, beautiful ways that millions of other people speak.

The big question everyone is asking is: Is this robot going to help everyone get their stories heard, or is it going to make everyone sound exactly the same, erasing the unique voices of the world? This is a huge deal because if we lose our different ways of speaking, we might lose different ways of thinking and understanding the world.

---

The Robot Librarian's Dilemma

A group of five language experts decided to sit down (virtually) and chat about this robot librarian. They called their conversation a "scholarly dialogue," but you can think of it as a roundtable of detectives trying to figure out if this new technology is a hero or a villain for linguistic diversity. They asked five big questions to see how Generative AI is changing the way scholars write and publish their work.

The Double-Edged Sword
The experts agreed that the robot is a "double-edged sword." On one side, it's a helpful tool. It can help a student who is nervous about their grammar get their ideas across clearly. It can help a researcher summarize a huge pile of books in seconds. It's like having a super-tutor that never gets tired.

But on the other side, the experts are worried. They found that the robot tends to make everything sound the same. It's like a photocopier that only knows how to print in black and white, even if you feed it a rainbow. If you ask the robot to write in a specific local style of English, like Nigerian English or Jamaican English, it often gets it wrong. Instead of writing a smart, academic paper in that style, it might accidentally turn it into a cartoonish version or a slang-heavy joke. It's as if the robot thinks, "Oh, you want Nigerian English? Here's some pidgin!" when the scholar actually wanted a serious, academic tone.

The "Standard" Trap
One of the biggest findings is that the robot is obsessed with one specific version of English: the American standard. Even if you ask it to write in British English, it often slips back into American spelling and grammar. It's like a DJ who only knows one song; no matter what genre you ask for, they keep playing the same hit.

The experts pointed out that this is dangerous. If scholars start using the robot to fix their writing, they might end up sounding like robots themselves. One young scholar mentioned that she is now afraid to use certain words like "delve" or "underscore" because she knows the robot loves them so much that if she uses them, people might think she is the robot! It's a weird world where humans are trying to sound less like machines to prove they are human.

The "Algorithmic Colonialism"
One of the experts, Sender, came up with a cool but scary term: "algorithmic colonialingualism." Imagine if a king came to your village and said, "Your way of speaking is messy. From now on, you must speak like us, or you can't trade." That's what the experts fear is happening with AI. The robot is built on data from the "Global North" (wealthy Western countries), so it doesn't understand the "Global South" (developing nations) or Indigenous ways of speaking. It's like trying to describe a tropical storm using only words for snow; the robot just doesn't get the context.

Who is in Charge?
The group also talked about who is responsible for fixing this. They said it's not just the scholars' fault; the universities and journals (the gatekeepers of knowledge) need to step up. Right now, many journals have vague rules about AI. Some say "don't use it," others say "use it but tell us." The experts suggest that journals need to be clearer and kinder. They need to tell their reviewers (the people who check the papers) that just because a paper doesn't sound like a textbook from London, it doesn't mean it's bad.

They also warned that if we let the robot do all the work, we might end up with a flood of papers that look perfect but say nothing new. It's like a factory churning out identical plastic flowers; they look nice, but they don't smell like real ones.

The Path Forward
So, what's the solution? The experts don't say we should throw the robot away. Instead, they say we need to use it carefully. They suggest that:

1. Humans must stay in charge: We should use the robot to help, not to write for us. We need to keep our own "voice."
2. Build better robots: The people who make these AI tools need to listen to people from all over the world, not just the US and UK. They need to train the robots on more diverse books and conversations.
3. Be honest: If you use the robot, you should say so. And if you are reviewing a paper, you shouldn't use the robot to judge it, because the robot might miss the point.

The Bottom Line
The paper concludes that the robot isn't inherently good or bad; it's just a tool. But right now, it's a tool that favors the powerful and the standard. If we don't pay attention, we might accidentally build a world where everyone thinks and sounds the same. The experts hope that by talking about this, we can steer the robot to be a tool for diversity, helping everyone's voice be heard, rather than a machine that silences the unique ones. It's up to us to make sure the robot learns to appreciate the whole library, not just the front page.