by 
A young woman is facing workplace trouble due to her reliance on OpenAI’s artificial intelligence (AI)-based chatbot, ChatGPT, which led to critical errors in her work. According to a Reddit post, the woman, employed in marketing and customer surveys, used ChatGPT to prepare and analyze data for an important client presentation. However, inaccuracies in the AI-generated output have put her job at risk.
The Reddit user, her boyfriend, shared that she used ChatGPT to create a customer survey questionnaire and later to analyze the responses. She uploaded the survey data into an Excel file, downloaded the AI-generated output, and presented it in a report. The issue arose during the presentation when she showcased results using the “Pearson’s Correlation Coefficient.” The survey responses, which were entirely text-based and categorized into five “feeling” types, were unsuitable for statistical analysis. ChatGPT attempted to convert the text into numerical data but failed to explain the process or provide detailed calculations, rendering the results flawed and unacceptable to the client.
