Can one trust ChatGPT? Hebrew U and Cambridge University mathematicians find out

ChatGPT generates responses by predicting sequences of words learned during its training. Now, a new Israeli study shows that ChatGPT’s unpredictability may limit its reliability in a math classroom.

THE EXPERIMENT, by two education researchers, asked the chatbot to solve a version of Plato’s slave-boy experiment of the ‘doubling the square’ problem.

(photo credit: WIKIPEDIA)

ByJUDY SIEGEL-ITZKOVICH

SEPTEMBER 28, 2025 01:29

Can one trust ChatGPT? Hebrew U and Cambridge University mathematicians find out

ChatGPT generates responses by predicting sequences of words learned during its training. Now, a new Israeli study shows that ChatGPT’s unpredictability may limit its reliability in a math classroom.

New dinosaur species identified from Welsh museum fossil

New research suggests media influences perception of aging

Can one trust ChatGPT? Hebrew U and Cambridge University mathematicians find out

ChatGPT generates responses by predicting sequences of words learned during its training. Now, a new Israeli study shows that ChatGPT’s unpredictability may limit its reliability in a math classroom.

See more on