Roman Lutz

Hilde Weerts, Miroslav Dudík, Richard Edgar et al. · microsoft-research

Fairlearn is an open source project to help practitioners assess and improve fairness of artificial intelligence (AI) systems. The associated Python library, also named fairlearn, supports evaluation of a model's output across affected populations and includes several algorithms for mitigating fairness issues. Grounded in the understanding that fairness is a sociotechnical challenge, the project integrates learning resources that aid practitioners in considering a system's broader societal context.

Ahmed Magooda, Alec Helyar, Kyle Jackson et al. · microsoft-research

We present a framework for the automated measurement of responsible AI (RAI) metrics for large language models (LLMs) and associated products and services. Our framework for automatically measuring harms from LLMs builds on existing technical and sociotechnical expertise and leverages the capabilities of state-of-the-art LLMs, such as GPT-4. We use this framework to run through several case studies investigating how different LLMs may violate a range of RAI-related principles. The framework may be employed alongside domain-specific sociotechnical expertise to create measurements for new harm areas in the future. By implementing this framework, we aim to enable more advanced harm measurement efforts and further the responsible use of LLMs.

Roman Lutz

As the shortcomings of our current Internet become more and more obvious, researchers have started creating alternative approaches for the Internet of the future. Their design goals are mainly content-orientation, security, support for mobility and cloud computing. The probably most popular architecture is called Content Centric Networking. Every communication is treated as a distribution of content and caches are used within the network to improve the effectiveness. While the performance gain of Content Centric Networks is undoubted, there are questions about security and especially privacy since it is not one of its main design principle. In this work, we compare the Content Centric Networking approach with the current Internet with respect to security and privacy. We analyze improvements that have been made and new problems that have yet to be resolved. The Internet of the future could be content-oriented, so it is essential to identify potential security and privacy issues that are inherent to the architecture early on.