7^th Feb 4PM

Ethical Aspects of Big Data and AI for Socio-political Challenges

Speaker: Tuba Bircan, Senior Social Scientists at Kavli Research Centre for Ethics, Science and the Public at University of Cambridge and Wellcome Connecting Science;  Research Professor of Sociology at the Dept. of Sociology, Free University of Brussels (VUB); Senior Research Fellow in the Population and Health Research Group at the University of St. Andrews

Abstract:

All spheres of life and social dynamics are beset with bias and various forms of discrimination. Numerous disadvantaged groups’ lacking- or mis-representation in Big Data and existing bias in Artificial Intelligence (AI) are no exceptions. AI algorithms learn by training and rely highly on “data”, therefore humans’ inherent biases in data and models are the main sources for the biases in AI. Unless the designs are improved and robust, AI algorithms are likely to reproduce and reinforce existing stereotypes and discriminatory social norms. Given the urgent and legitimate demand for ethically responsible, explainable and fair AI; the ethical aspects of Big Data are shaped by concerns regarding the social consequences of both the widespread adoption of AI and the ways in which biases in data can be replicated and perpetuated. This seminar will focus on the discussing following questions to assess the obstacles and challenges for developing ethical, fair and vulnerability-responsive AI: What causes biases in AI? What are the examples from real-life AI applications for socio-political challenges? What are the possible approaches/practices to avoid bias in AI?

Speaker’s bio:

Dr. Tuba Bircan is a senior social scientists at Kavli Research Centre for Ethics, Science and the Public at University of Cambridge and Wellcome Connecting Science, besides her appointment as a research professor of sociology at the Dept. of Sociology, at the Free University of Brussels (VUB). As an interdisciplinary computational social scientist, her research interests cover a wide range from migration, inequalities, social and public policies to new methodologies and use of Big Data and AI for studying socio-political challenges. She is a follower of open science and science for society. She is currently leading the H2020 funded Enhanced Migration Measures from a Multidimensional Perspective (HumMingBird) project and is the co-promoter of Climate-Induced Migration in Africa and Beyond: Big Data and Predictive Analytics (CLIMB) project. She sits in editorial board of Nature Humanities and Social Sciences Communications and Plos One, as well as ethical board of AI Excellence Centre of Flemish Employment Agency in Belgium. She has published mainly on the methodological aspects of migration studies, use of Big  Data and AI in social sciences and inclusive policies and inequalities.

Zoom link to attend.
Meeting ID: 394 099 2896
Passcode: SOST

Related