Detecting Computer Generated Text


Harvard University researchers developing method that identifies computer generated text. The team’s central tenant was whether a system can be built to detect generated text

Using that idea, Gehrmann and Strobelt developed a method that, instead of flagging errors in text, identifies text that is too predictable.

“The idea we had is that as models get better and better, they go from definitely worse than humans, which is detectable, to as good as or better than humans, which may be hard to detect with conventional approaches,” said Gehrmann.

Harvard Gazette