To See or Not to See - Is AI at Its Best When People Know It's There?
Offered By: Data Science Festival via YouTube
Course Description
Overview
Explore a thought-provoking panel discussion from the Data Science Festival examining the transparency of AI in user-facing services. Delve into the crucial question of whether AI systems should be openly disclosed or kept hidden from users. Compare approaches across platforms like Reddit's open-source algorithms, opaque online shopping relevance sorting, and Google's semi-transparent search engine. Join speakers Kieron Maddison, Sarah Ying Zheng, Tessa Darbyshire, and Walter Pasquarelli as they debate the pros and cons of AI visibility and its impact on user experience and trust. Gain insights into the ethical considerations and strategic decisions behind AI transparency in various digital services.
Syllabus
To see or not to see – is AI at its best when people know it’s there (and when it’s not)?
Taught by
Data Science Festival
Related Courses
Calvin - Histoire et réception d'une RéformeUniversity of Geneva via Coursera Moralities of Everyday Life
Yale University via Coursera Introduction to Bioethics
Georgetown University via edX Les problèmes métaphysiques à l’épreuve de la politique, 1943-1968
École normale supérieure via Coursera Business Ethics for the Real World
Santa Clara University via Canvas Network