Experiment ended

Abstract

In this article, we present the outcome of our first experiment with Chatbase, a chatbot built on chatGPT’s functioning model(s). Our idea is to try instructing Chatbase to perform a reading, digesting, and summarizing task for a specifically formatted academic document.

Experiment ended

“…Its answer to Prompt 5 is a little bit mechanic. And interestingly, similar to its suggestions in [1] for the last question, the AI provides all the non-existent document titles. Nonetheless, it is noteworthy that Chatbase has the capability to identify notable authors writing about the topic, including the co-founders of Retraction Watch, Ivan Oransky and Adam Marcus.…”

“…

In this second experiment, we updated the data for the chatbot provided in [1] by uploading a new article [2] to the Chatbase server for learning.Still, we would like to explore Chatbase's reading comprehension skills as well as its capabilities in delivering summarized or specifically extracted content from the provided article [2]. In addition, in our prompts given to Chatbase concerning the content of [2], we tried to learn if the AI could relate the read content to principles in [3][4].We provide detailed conversations with Chatbase in what follows.Prompt 1: What are the key lessons learned from this article?Chatbase: The article discusses the importance of retractions in scientific research and suggests ways to improve the retraction process.

…”

“…In this second experiment, we updated the data for the chatbot provided in [1] by uploading a new article [2] to the Chatbase server for learning.…”