People working on this Project: Brianna Hawkins, Roza Danaeifard, T’Neil Gooden, Isaac Chabot, and Karen Farkas.
Our project is about understanding how accurate the Theory of Mind is in Large Language Models (Artificial Intelligence). We used ChatGPT 3.5 and 15 adult humans to identify how the Theory of Mind is viewed and compared in Humans and Artificial Intelligence.
We had four different topics to choose from in our Seminar and we all decided that the most interesting one to us was how ChatGPT 3.5 and Humans compared when understanding ToM. We used five different Theory of Mind prompts on both ChatGPT and Humans to see how well both elements would answer the questions.
When we asked both ChatGPT 3.5 and Humans the prompts we used Excel sheets to copy the data and create a results chart from that data. We were able to determine the answer to our hypothesis from this data and that is what we will be presenting on.
