OpenAI ChatGPT, Google Bard, and Microsoft Bing: Similarity Index and Analysis of Artificial Intelligence-Based Contents

Abstract

This research compares AI language models—specifically, OpenAI's ChatGPT, Google's Bard, and Microsoft's Bing—by analyzing their originality, use of external sources, and performance. Conducted on August 5, 2023, the study evaluates how these models respond to different queries, revealing distinct characteristics. ChatGPT stands out with lower similarity scores and lesser reliance on online sources, indicating its potential for creating more unique content. In contrast, Bard and Bing show higher similarity scores, suggesting they draw more from available online content, which could be beneficial for tasks requiring context-rich information. While ChatGPT and Bard excel in grasping context, generating substantial content, and offering relevant insights, there are concerns about accuracy and consistency across queries. Notably, Bing's focus on aiding content creation rather than direct essay generation showcases diverse strengths among AI models. As AI technology progresses, refining these models and addressing inconsistencies will improve their usefulness across various applications. These findings guide users in choosing AI tools that fit their content needs and ensure the credibility of generated outputs.