ChatGPT, an AI system by OpenAI, is making waves in the artificial intelligence field with its advanced language capabilities. Capable of performing tasks such as drafting emails, conducting research, and providing detailed information, such tools are transforming the way office tasks are conducted. They contribute to more efficient and productive workplaces. As with any technological advancement, understanding user interaction and behavior with AI systems like ChatGPT is crucial for their broader adoption.
Traditional studies exploring this interaction have largely focused on user satisfaction metrics and system functionality. They often involve analyzing user feedback, performance benchmarks, and carrying out usability studies. However, these evaluation methods often fail to analyze how unique AI characteristics impact user behavior and knowledge transfer.
In response to this gap in research, a joint study conducted by Kookmin University and HJ Institute of Technology and Management aimed to examine how the system updates, memorability, and multilingual capabilities of ChatGPT influence knowledge acquisition, application, and user satisfaction.
The research involved data collection from experienced ChatGPT users, followed by analysis via structural equation modeling (SEM). The findings showed that system updates greatly enhanced ChatGPT’s ability to provide accurate and extensive information, making it more efficient for users to gain and apply knowledge. A path coefficient of 0.41 for knowledge acquisition and 0.42 for knowledge application was observed.
Similarly, the memorability characteristic of ChatGPT positively influences knowledge acquisition (0.19) and application (0.25), thanks to its ability to recall past interactions and user preferences.
The study also showed the importance of overcoming non-language barriers through multilingual support, especially in diverse workplace environments. Non-language barriers were found to influence knowledge acquisition (0.18) and application (0.19).
Another area of focus was ChatGPT’s human-like attributes and their impact on user satisfaction and word-of-mouth (WOM) recommendations. The findings showed that these traits significantly enhance user satisfaction (0.16) and word-of-mouth recommendations (0.39).
In conclusion, system updates, memorability, and non-language attributes enhance the effectiveness of AI tools like ChatGPT. By improving knowledge acquisition and application, increasing user satisfaction, and promoting positive WOM recommendations, these features contribute heavily to productivity and engagement in the workplace. These insights serve to guide the development of AI tools towards more usefulness and broader acceptance among users.
