The Responsible AI Symposium brought together more than 300 students and guests from the university and industry. The two-day programme covered practical and research questions around the use of AI systems.
Day 1: AI Governance
The first day opened with a keynote on AI governance and large language models. The discussion covered fairness, transparency and accountability when AI systems are used in real settings.
A panel on bias in machine learning looked at training data, human review and ways to test models before they are used. Researchers and industry speakers shared examples from their work.
Day 2: Practical Sessions
The second day focused on workshops. Participants worked with real datasets and tried techniques for finding and reducing bias in machine learning models. The sessions included adversarial debiasing, calibrated equalized odds and fairness constraints.
Student Hackathon
A six-hour hackathon asked teams to build prototypes related to responsible AI. The winning team built a tool that checked hiring models for differences in candidate scores across demographic groups.
After the Event
The chapter plans to continue the topic through workshops and reading sessions on AI ethics and responsible computing.