This 22-year-old is trying to save us from ChatGPT before it changes writing forever
While many Americans were nursing hangovers on New Year's Day, 22-year-old Edward Tian was working feverishly on a new app to combat misuse of a powerful, new artificial intelligence tool called ChatGPT.
Given the buzz it's created, there's a good chance you've heard about ChatGPT. It's an interactive chatbot powered by machine learning. The technology has basically devoured the entire Internet, reading the collective works of humanity and learning patterns in language that it can recreate. All you have to do is give it a prompt, and ChatGPT can do an endless array of things: write a story in a particular style, answer a question, explain a concept, compose an email — write a college essay — and it will spit out coherent, seemingly human-written text in seconds.
The technology is both awesome — and terrifying.
"I think we're absolutely at an inflection point," Tian says. "This technology is incredible. I do believe it's the future. But, at the same time, it's like we're opening Pandora's Box. And we need safeguards to adopt it responsibly."
Tian is a senior at Princeton University, where he majors in computer science and minors in journalism. Before his recent foray into the limelight, Tian's biggest plans were graduating college and getting his wisdom teeth pulled. Now he's fielding calls from venture capital firms, education leaders, and global media outlets.
Over the last couple years, Tian has been studying an AI system called GPT-3, a predecessor to ChatGPT that was less user-friendly and largely inaccessible to the general public because it was behind a paywall. As part of his studies this fall semester, Tian researched how to detect text written by the AI system while working at Princeton's Natural Language Processing Lab.
Then, as the semester was coming to a close, OpenAI, the company behind GPT-3 and other AI tools, released ChatGPT to the public for free. For the millions of people around the world who have used it since, interacting with the technology has been like getting a peek into the future; a future that not too long ago would have seemed like science fiction.
Despite having studied AI, Tian, like the rest of us, was gobsmacked by the power of ChatGPT. He and his friends used it to write poems and raps about each other. "And it was like: 'Wow, these results are pretty good,'" Tian says. It seemed like everyone on campus was talking about how remarkable this new technology was. Sure, the text it generates is pretty formulaic and not always accurate. But it also feels like the beginning of a revolution.
For many users of the new technology, wonderment quickly turned to alarm. How many jobs will this kill? Will this empower nefarious actors and further corrupt our public discourse? How will this disrupt our education system? What is the point of learning to write essays at school when AI — which is expected to get exponentially better in the near future — can do that for us?
Stephen Marche, writing in The Atlantic last month, declared "The College Essay Is Dead." He paints ChatGPT and the AI revolution as part of an existential crisis for the humanities. "The essay, in particular the undergraduate essay, has been the center of humanistic pedagogy for generations," Marche writes. "It is the way we teach children how to research, think, and write. That entire tradition is about to be disrupted from the ground up."
Edward vs The Machine
After the fall semester ended, Tian traveled home to Toronto for the holidays. He hung out with his family. He watched Netflix. But he couldn't shake thoughts about the monumental challenges confronting humanity due to rapidly advancing AI.
And then he had an idea. What if he applied what he had learned at school over the last couple years to help the public identify whether something has been written by a machine?
Tian already had the know-how and even the software on his laptop to create such a program. Ironically, this software, called GitHub Co-Pilot, is powered by GPT-3. With its assistance, Tian was able to create a new app within three days. It's a testament to the power of this technology to make us more productive.
On January 2nd, Tian released his app. He named it GPTZero. It basically uses ChatGPT against itself, checking whether "there's zero involvement or a lot of involvement" of the AI system in creating a given text.
When Tian went to bed that night, he didn't expect much for his app. "When I put this out there, I just thought maybe a few dozen people at best might try it," Tian says. "I was not expecting what happened."
When Tian woke up, his phone had blown up. He saw countless texts and DMs from journalists, principals, teachers, you name it, from places as far away as France and Switzerland. His app, which is hosted by a free platform, became so popular it crashed. Excited by the popularity and purpose of his app, the hosting platform has since granted Tian the resources needed to scale the app's services to a mass audience.
Fighting The Hallmarkization Of Everything
Tian says he has a couple primary motivations for creating GPTZero. The first is transparency. "Humans deserve to know when something is written by a human or written by a machine," he says.
Along these lines, one obvious application for GPTZero is to help teachers identify whether their students are plagiarizing their essays from ChatGPT. "Teachers from all over the world are worried about this," Tian says.
Some in the technology world, however, are not quite sold that copying and pasting what ChatGPT spits out is even a problem. "'ChatGPT plagiarism,' is a complete non-issue," tweeted Marc Andreessen, a venture capitalist and Internet pioneer, earlier this month. "If you can't out-write a machine, what are you doing writing?"
Elon Musk, one of the original co-founders of OpenAI, recently tweeted, "It's a new world. Goodbye homework!" in response to reports that schools were imposing strict new measures against ChatGPT plagiarism.
Of course, these are just flippant tweets. But it really does feel like we've entered a new world where we're being forced to re-evaluate our education system and even the value — or at least the method — of teaching kids how to write.
Many of us lost our will — even our ability — to remember phone numbers when cell phones came along. By outsourcing memorization to a machine, we've become dependent on it to call our friends and family. You might say it's been for the best, and it's freed our minds to concentrate on other matters. Or you might consider it a kind of de-evolution, a dumbing down of our mental abilities. Don't lose your cell phone!
Now humanity faces the prospect of an even greater dependence on machines. It's possible we're heading towards a world where an even larger swath of the populace loses their ability to write well. It's a world in which all of our written communication might become like a Hallmark card, written without our own creativity, personality, ideas, emotions, or idiosyncrasies. Call it the Hallmarkization of everything.
But at least when we give people Hallmark cards, people know we're giving them Hallmark cards. If you use ChatGPT to write your friend a congratulations or an apology, they might not even know it was written by a machine.
Which brings us to the other purpose that Tian envisions for his app: to identify and incentivize originality in human writing. "We're losing that individuality if we stop teaching writing at schools," Tian says. "Human writing can be so beautiful, and there are aspects of it that computers should never co-opt. And it feels like that might be at risk if everybody is using ChatGPT to write."
Tian is no Luddite. He isn't trying to stop AI in its tracks. He believes that's impossible, and, he says, he opposes blanket bans against use of ChatGPT, like the one recently announced by New York City public schools. Students, he believes, will use the technology anyway. And, he says, it's important they're able to learn how to use it. They need to be aware of the technological changes that are sweeping our world. "It doesn't make sense that we go into that future blindly," he says. "Instead, you need to build the safeguards to enter that future."
As for his plans after college, Tian says, the excitement — and clear demand — for his new app has convinced him that he should concentrate on making it a better, more accurate product. "If you're a teacher or an educator, our team — which right now is just me and my best friend from college, who just joined yesterday — we would love to talk to you," Tian says.
So if you encounter some text that you suspect may be written by a machine, maybe run it through Tian's new app? You can find it at GPTZero.me.
Copyright 2023 NPR. To see more, visit https://www.npr.org.