In the evaluation conducted by philosopher and author Eliette Abécassis and teacher Lev Frankel, “Chat GPT” received a grade of 11/20, while Raphaël Enthoven received a perfect score of 20/20. This grading process took place during an experiment held in Paris on Wednesday.
The program “Chat GPT” was specifically designed to generate a philosophical article and was trained accordingly. In the experiment, it produced a copy of a philosophical topic, which had previously received a passing grade in the Baccalaureate stage in France. The grading responsibility fell upon Eliette Abécassis and Lev Frankel, also known as “Cyril Thinker” on TikTok.
The chosen topic for the program to write about was whether joy can be considered a rational matter. This topic served as the basis for an intellectual showdown between “Chat GPT” and the philosopher Raphaël Enthoven, organized by a specialized school of commerce and technology.
According to Eliette Abécassis, the evaluation of the two versions immediately revealed the distinctive writing style of each author from the very first words.
Lev Frankel expressed his criticisms of the “Chat GPT” version, stating that it lacked any significant problems. He criticized the article for consisting of excessively long sentences with little substance, making it difficult for the reader to understand the arguments presented. Additionally, Frankel noted the excessive use of quotations in an attempt to compensate for the lack of meaningful content.
Frankel further commented that the material produced by “Chat GPT” did not truly embody philosophy, as it failed to deliver beautifully constructed sentences. He also pointed out numerous errors in the section referencing the authors, indicating its overall weakness.
