AI's risks and biases ft. Timnit Gebru

Presidential Aiken Lecture

Timnit Gebru is Founder and Executive Director of the Distributed AI Research Institute (DAIR) and was named one of TIME Magazine's 100 Most Influential People (2022). She earned her PhD in Computer Science from Stanford University under the advisement of Fei-Fei Li. Co-founder of Black in AI (2017) and the ACM FAccT Conference (Fairness, Accountability, and Transparency in Computing), she was formerly Co-Lead of Google's Ethical AI Team before her controversial departure in 2020 after refusing to retract a research paper on AI risks, research that has since proven prescient about large language model concerns. Her "Gender Shades" research, which won the VentureBeat AI Innovation Award, proved that commercial facial recognition systems fail on Black women at 35% error rates. With 20,000+ citations, Gebru demonstrates extraordinary moral courage, having sacrificed her dream position at Google to speak truth about AI's risks and biases. She now runs an independent research institute centering African and diaspora communities in AI development, showing students that AI ethics isn't abstract philosophy, it's about real people being harmed by biased systems right now.