Assessing ChatGPT's Efficacy in Interpreting Privacy Policies

In the contemporary digital landscape, the intricacy of privacy policies poses a considerable challenge, often leaving users confused and at risk of compromising their privacy. This study explores the capabilities of generative AI, with a particular focus on ChatGPT-4, to aid users in deciphering, summarizing, and interpreting an array of 102 privacy policies. Utilizing the Flesch-Caid readability score as a benchmark, our findings reveal that ChatGPT enhanced the readability of 56 out of the 102 policies evaluated. However, in certain instances, the interpretations by ChatGPT resulted in reduced readability (such as–), highlighting potential limitations of the current state of AI in processing complex legal text. Despite this, the application of ChatGPT’s capabilities is seen as a pivotal step towards distilling complex legal verbiage into explanations that are accessible with lesser complex jargon, thereby enabling users to make well-informed decisions.