Organizations can use computers or AI to make decisions about people: digital differentiation. For example, insurers can adjust prices to consumers, and the government can use AI-driven analysis to combat welfare fraud. Such digital differentiation is often useful and efficient, but it also brings discrimination-related risks. First, there is a risk of discrimination against people with a certain ethnicity, gender, or similar characteristics. Second, there is a risk of other unfair differentiation that does not specifically affect people with a particular ethnicity or similar characteristic, but is still unfair. For example, digital differentiation can reinforce economic inequality. The presentation introduces the main applicable rules in Europe, such as non-discrimination law and in the General Data Protection Regulation (GDPR). The presentation also shows that those rules, while useful, leave serious gaps. Bio Frederik Zuiderveen Borgesius is a professor of ICT and law. He works at t
As part of the CITP Digital Investigators Conference, the public is invited to join us in person or virtually for this event. Please register here to attend in person. The livestream will be available here. Powerful new technologies like OpenAI’s “ChatGPT” or Google’s “Bard” have sparked excitement over the potential they have to transform how we work, learn and communicate for the better. But their potential harms also trigger fears and unease. As a result, the public discourse around such large language models (LLMs) can be noisy or chaotic. CITP has convened a panel of experts from the journalism, tech research and public policy sectors to discuss their experiences with – and approaches to – engaging with these emerging technologies in their respective professions. We will also talk about the responsibilities journalists and academics may have in shaping the public conversation around digital technologies, and how they can support each other’s work for the b
Incorporating ethics and responsibility explicitly into tech teams’ workflow meaningfully is an industry-wide challenge. However, if done well, it has the potential to transform which technologies are deployed in society, and how. We have found that technologists are generally eager to address the ethical and responsibility dimensions of their work but frequently encounter frustration and swirl due to a lack of shared vocabulary, a knowledge of relevant frameworks, time, and a dedicated forum to have discussions. We have developed a methodology called the “Moral Imagination Workshop” to help teams learn about, think through, and apply the tools and methodologies of ethics and responsibility as it relates to their work. In this talk, Ben will share our motivations, aims, and some experiences from the Moral Imagination Workshop methodology for technical teams. This talk speaks to one of Google’s efforts toward responsible innovation. Bio: Ben Zevenbergen works at Google as a “
Machine learning algorithms are widely used for decision making in societally high-stakes settings from child welfare and criminal justice to healthcare and consumer lending. Recent history has illuminated numerous examples where these algorithms proved unreliable or inequitable. This talk will show how causal inference enables us to more reliably evaluate such algorithms’ performance and equity implications. In the first part of the talk, it will be demonstrated that standard evaluation procedures fail to address missing data and as a result, often produce invalid assessments of algorithmic performance. A new evaluation framework is proposed that addresses missing data by using counterfactual techniques to estimate unknown outcomes. Using this framework, we propose counterfactual analogues of common predictive performance and algorithmic fairness metrics that are tailored to decision-making settings. We provide double machine learning-style estimators for these metrics that achieve