All-knowing chatbots took the internet by storm last year, convincing one engineer that machines have become sentient, and creating fears of an epidemic of cheating in schools.
ChatGPT in education
Alarm among educators has reached a fever pitch in recent weeks over ChatGPT, an easy-to-use artificial intelligence (AI) tool trained on billions of words and tons of data from the web.
Being able to write a half-decent essay and answer a lot of common classroom questions, sparks a fierce debate about the future of traditional education.
Also read: ChatGPT takes the world by storm – viral AI bots explained
Pandemic lies
The New York City department of education banned ChatGPT from its network due to “concerns about the negative impact on student learning”.
“While these tools can provide quick and easy answers to questions, they don’t build critical thinking and problem-solving skills,” said Jenna Lyle of the department.
A group of Australian universities has said it will change its exam format to eliminate AI tools, which it sees as outright cheating.
Opportunity or threat?
However, some in the education sector are more relaxed about AI tools in the classroom, and some even perceive them as an opportunity rather than a threat.
This is partly because ChatGPT in its current form is still flawed.
To give an example, think Guatemala is bigger than Honduras. It is not.
Also, ambiguous questions can throw off the track.
Ask the tool to describe the Battle of Amiens and it will provide a detail or two about the 1918 confrontation of World War I.
ChatGPT is an ‘important innovation’
But not the flag that there was also a skirmish of the same name in 1870. It takes some guidance to realize the mistake.
“ChatGPT is an important innovation, but no more than a calculator or a text editor,” said French author and educator Antonio Casilli. AFP.
“ChatGPT can help people who are stuck with a blank piece of paper to write their first draft, but after that they still need to write and style.”
Also Read: ChatGPT: ‘Professional’ version of human-like chatbot is coming
Allow ChatGPT, within reason
Researcher Olivier Ertzscheid from the University of Nantes agrees that teachers should focus on the positive.
In any case, he told me AFPHigh school students have been using ChatGPT, and any attempt to ban it will only make it more attractive.
Teachers should “experiment with the limits” of AI tools, he said, by generating texts themselves and analyzing the results with students.
Beware of plagiarism
Universities already use software that detects plagiarism, so it doesn’t take much imagination to see a future where every essay is run through an AI detector.
The campaign also floated the idea of a digital watermark or other form of signature that would identify AI work and end the ‘fraud pandemic’.
And OpenAI, the company that owns ChatGPT, says it’s working on a “statistical watermark” prototype.
This shows that educators will do well in the long run.
A changing landscape
Belying the fear of the pandemic, Casilli, for one, still believes the impact of the tool has a great symbolic meaning.
He said that some of them broke the rules of the game, so the teacher questioned his students.
Now, students ask the machine before checking everything in the output.
“Every time a new tool appears, we worry about potential abuse, but we also find ways to use it in our teaching,” Casilli said.
Jules Bonnard © Agence France-Presse