This workshop allows participants to explore (new) design methodologies for developing ethically sound AI that takes account of its societal context. Participants will work in small teams. Each team looks at two concrete case studies based on real life AI projects. For each project the teams will explore the Human Centered Design assumptions behind the project, how these assumptions focus on the individual user and what problems might come from that focus, and how a more collective perspective (i.e., seeing users as members of society) could have led to different outcomes. The teams will then connect with another team to see if they can come up with a set of changes that we need to make to our standard design methodology. These changes come in the form of: “Rather than ..X.. we will now do ..Y..”.
This workshop used two worksheets (part 1 and part 2) and let to these results.
There are two goals/intended outcomes for our session: firstly, we want to facilitate the thinking process around design methodologies for ethical AI for the participants of the workshop (through bringing them case studies and a smart workshop guide which invites thought). Secondly, we hope to learn from the participants in order to further our own thinking around what would be a proper methodology for designing AI for the collective good.
We're hoping that many efforts and discussions will continue after Mozfest. Share any ideas you already have for how to continue the work from your session.:We aim to continue the work in three ways: Firstly, we will summarize the outcomes of the workshops in an easy to read and accessible one-pager and in a complete overview of the raw (anonymized) output, which we will share with the participants of the workshop and make public under a permissive Creative Commons license. Secondly, we will offer to host a small mailing list for the people who want to stay connected and continue to work on new methodologies for the development of AI technology. Finally, the theme of this workshop is at the core of our organization, so we will thus use the outcomes of the session directly in furthering our public work.
How will you deal with varying numbers of participants in your session?:The workshop will mostly consist of people doing work in small teams (most likely of three people each). We will prepare the workshop in such a way that it will require very little facilitation (the teams will guide themselves), and will have the small groups present to each other in order to learn from each other. This allows the workshop to flexibly scale from a very small number to a very large number of participants (limited by the usability of the breakout-room settings of the conference software). Only once we collate the work of the different groups will we be able to see the true outcome of the workshop.
Hans is currently a lecturer/researcher. As the former director of digital rights organization Bits of Freedom and as a philosopher, he has an interest in the politics of tech.
