60% of the responses from the GPT-3.5 neural network model are plagiarized, a Copyleaks study showed. The greatest plagiarism is found in the answers to physics and chemistry questions: they are paraphrased fragments and repetitions.
Approximately 60% of the responses from the GPT-3.5 neural network are plagiarized. This is stated in the results of the Copyleaks study.
To carry out the study, analysts generated more than a thousand texts using GPT-3.5 on various topics: physics, social studies, sports, philosophy, sociology and others.
59.7% of the neural network responses contained plagiarism. Most of the answers turned out to be repetitions and paraphrased elements. Researchers noted the highest percentage of plagiarism to one degree or another in answers to chemistry and physics questions.
Copyleaks emphasized that neural network models are not capable of creating new things, but can only transform existing information.
Author:
Kirill Bilyk
Source: RB
I am a professional journalist and content creator with extensive experience writing for news websites. I currently work as an author at Gadget Onus, where I specialize in covering hot news topics. My written pieces have been published on some of the biggest media outlets around the world, including The Guardian and BBC News.
