Asking right, reading tone: a DuPont experiment on Artificial Intelligence's performance.

The objective of this thesis is to identify the most effective variables and the optimal sequences for their input, in order to make the AI more accurate and efficient while minimizing data misinterpretations, within the framework of DuPont analysis for evaluating business performance. This experiment is repeated across different types of AI to assess potential differences and determine which models are the most effective. A subsequent analysis will examine the financial tone, word frequency, and the degree of optimism or pessimism used by the various artificial intelligence systems.