There are now “serious concerns” children will use ChatGPT artificial intelligence to cheat on their homework, according to Jess Kelly.
The Newstalk tech correspondent said the latest version of the AI chatbot is the most intelligent to date – and could write you up an “incredible” essay on almost anything in a very short space of time.
She said schools in the US are now very concerned about the technology being misused for homework – with the New York Department of education already moving to ban the technology from public school devices and networks.
“A lot of other places in the US now are looking to implement similar protocols because there are fears that this technology is so good, it could fool the professors,” Jess told Newstalk Breakfast.
She said the latest version of ChatGPT is capable of providing users with information in a “very creative, very natural way that is very easy to understand”.
“It will write an incredible essay for you,” she said.
“Not only does it have the data of the particular topic you are talking about, it now has that ability to write something in a really conversational way – and if it’s too conversational, you can go back saying, ‘can you put this in a more formal tone,’ for example.
“So say you’re in your English class and you’ve been asked to write 1,000 words on the themes of Pride and Prejudice.
“You can quite literally go to Chat GPT, put that in and you will get back a 1,000-word essay that is stunning to read, factually accurate for the most part and then you can submit it as your own work.”
Asked whether the situation could see schools going back to the old days of pen and paper, Jess said: “I don’t know how they could possibly do that – particularly with homework and assignments and so on”.
“I do think we are going to see rules and regulations around it,” she said.
“It is not just going to be education that this impacts. You could read an interview in a paper with a politician or watch a politician deliver an address on television or radio and that could have been written by ChatGPT.
“This technology is so sophisticated.
“I do think in schools they will have to introduce some form of policies around it but I do think it is going to be a difficult thing to police.”
Jess said the technology will need regulation to prevent people from misusing it.
“Although it is very smart, it doesn’t know everything and it can be manipulated,” she said.
“That is the other fear factor with this – that it can almost be weaponised to put out misinformation so although it is brilliant, I don’t think it is the be all and end all.”
You can listen back here: