Schools are unprepared for ChatGPT’s impact on teaching and learning, computer science teachers said.
A study conducted by BCS, the Chartered Institute for IT, found that 62 percent of professionals believe chatbots like ChatGPT make it harder to fairly evaluate student work.
ChatGPT is a Large Language Model (LLM) that can answer questions in a seemingly natural way and is trained on a huge dataset. It’s shown to be capable of producing university-level passing answers — including passing university law exams — but it’s fallible. A recent public demo of Google’s own AI service, Bard, provided an incorrect answer.
The majority (56 percent) of the 124 computer science teachers in the BCS study did not believe their school had a plan to manage incoming student use of ChatGPT, while 33 percent said early discussions had taken place, and another 11 percent said they did Plan was made.
Over three quarters of computer teachers (78 percent) rated the overall awareness of the potential of ChatGPT among peers at their school or college as “low” or “very low”.
However, a recent study found that students using the software to complete essay assignments could risk plagiarism due to the way the AI processes text.
45 percent of computer teachers were confident that ChatGPT is a tool that will improve teaching at their school in the long term by helping with assignment planning and assisting students with research techniques.
Julia Adamson, executive director of education and nonprofits at BCS, said: “Assuming these generative AI programs remain open-ended, teenagers will use them to answer homework just as adults will rely on them for work.
“Computer teachers want their peers to use AI as a great way to enhance learning in the classroom. However, they believe that without the right technical tools and guidance, schools will struggle to help students evaluate the responses they receive from chatbots.
“Calculators used to be forbidden in exams, now they are mandatory. We need to introduce machine learning into mainstream classroom practice, otherwise children will use AI unsupervised for homework without understanding what it is telling them.
“Another danger is that the digital divide will only widen if better-off parents can pay for premium chatbot services — and get better answers.”
Concerns have already been raised about how chatbots could spread misinformation due to their inability to understand the authenticity of their sources and the broader implications of technology’s impact on writing-based professions.
Sign up for the E&T News email to get great stories like this in your inbox every day.
source link
