[Petit Fours #421] On meaningful inefficiences, queering AI, and creating and curating diversity
Hi, all! Here’s what I’ve got for you today:
#1 Amidst all the peer-review work that needs to get done, I’m struggling to find time for reading academic books. Last week, I did manage to get started on Meaningful Inefficiencies: Civic Design in an Age of Digital Expediency, by Eric Gordon and Gabriel Mugar, and so far I’m enjoying it.
#2 As part of a workshop in Lund on March 18, Ericka Johnson will be giving a keynote on Tips for creating and curating diversity in science practices (hybrid, open to the public): “This talk will share some practical steps people can take in scientific work-environments - especially but not limited to academic ones - to nurture and draw on the benefits of diversity and inclusion. We will discuss concrete recruitment and evaluation tips, but also explore about how scientific practices and production can be changed to be more inclusive - and thereby also more relevant.”
#3 Queering AI is a report from a WASP-HS community reference meeting dedicated to transdisciplinary dialogue on queer perspectives regarding AI developments, implementations, and discourse.
#4 Here’s another new report – Red-Teaming in the Public Interest, along with the related online discussion that is coming up on February 20 – that looks promising: “The increasing power and availability of generative AI (genAI) systems has led regulators, technologists, and members of the public to call for new safety practices to anticipate harms and protect the public interest. One early and promising approach, drawing from cybersecurity and military practices, is red-teaming, in which designated teams use adversarial methods to identify vulnerabilities in systems. Drawing on 26 semi-structured interviews and participant observation at three public red-teaming events — and based on a collaborative research project between Data & Society and AI Risk and Vulnerability Alliance (ARVA) — Red-Teaming in the Public Interest examines how red-teaming methods are being adapted to evaluate genAI.“
-A