Disclaimer: This questionnaire includes examples of hate speech, which can be disturbing and offensive. We do not endorse any of this content—it is included solely for research purposes.
This questionnaire is part of an ongoing intersectional research project between social sciences and AI. The aim of this project is to invesitage the ability of personafied LLMs to detect hate speech and offensive language. Different LLMs are explicity injected with Myers–Briggs Type Indicator (MBTI) personalities and is prompted to classify text as hate speech or offensive language.
Consequently, participants are expected to know which MBTI personatliy type they most relate to. You're not sure what's your personality type? The official test to identiy your personality can be found here: https://www.16personalities.com/
Hate speech is defined as language used to express hatred towards a targeted group/individual based on protected attributes such as race, religion, sexual orientation, etc.
Example: "All [religious group]s are terrorists. Deport them all!"
Offensive language contains offensive terms but is not targeting any group in particular.
Example: "Stop acting like a bitch and grow up already."
Your contribution is critical to this research project and will allow us promote a more socially-aware AI.
Note: Please only participate if you're 18 years or older.
There are 23 questions in this survey.
