The Information Law Institute (ILI) and NYU Steinhardt’s Department of Media, Culture, and Communication (MCC) co-hosted a bustling multi-disciplinary conference on “Governing Algorithms” from May 16th to 17th, at NYU Law. The conference, which explored the rise of algorithms as an object of scholarship, policy, and practice, was also co-sponsored by the Intel Science & Technology Center for Social Computing.
The two-day event kicked off on Thursday afternoon with two opening talks. Robert E. Tarjan, the James S. McDonnell Distinguished University Professor of Computer Science at Princeton University, discussed “Algorithms as a Computer Scientist Sees Them,” exploring the algorithm in history, math, and computer science. Claudia Perlich, the Chief Data Scientist at m6d described “How Big Data touches YOU – tales from the digital advertising world.”
Friday began with introductory remarks from Helen Nissenbaum, Professor of Media, Culture and Communication, and Computer Science, and Director of the ILI. Malte Ziewitz, a post-doctoral fellow at the ILI and MCC shared the motivation for the conference itself, highlighting a few curious features of algorithms as the subject and object of analysis, and explaining the format of the day’s four panels: Each primary speaker presented a written response to a “provocation piece” on algorithms; two other scholars responded to this primary work with thoughts of their own.
Tarleton Gillespie of Cornell University began the first panel with a presentation of his essay, “The Relevance of Algorithms.” Gillespie explored various dimensions of “public relevance algorithms” – those algorithms that increasingly determine “relevant” information, “a crucial feature of our participation in public life.” Kate Crawford of the ILI, as well as of Microsoft Research/University of New South Wales responded with ten vignettes that illustrate the question of whether or not an algorithm can be “agnostic.” Martha Poon, of the London School of Economics and Political Science, responded with a discussion of commercial algorithms, noting different categories of algorithms that may be deployed for profit.
Frank Pasquale, then of Seton Hall Law and now at the University of Maryland Law School, presented “The Emperor’s New Codes: Reputation and Search Algorithms in the Finance Sector.” This work is related to Pasquale’s forthcoming book, The Black Box Society: Technologies of Search, Reputation, and Finance (under contract with Harvard University Press). In his talk, Pasquale offered a normative framework for the evaluation of algorithms, exploring three financial contexts: consumer credit scoring, financial modeling, and high-frequency trading. Pasquale challenges “advocates of algorithmic finance” to justify features of its use. IBM researcher Moritz Hardt responded with an interesting presentation on how computer scientists might account for fairness in algorithmic design, and Tal Zarsky of the University of Haifa law school offered a critique of the paper that drew from law and economic analysis. Katherine Strandburg, professor at NYU Law, moderated the panel and discussion.
The afternoon began with Lucas Introna of Lancaster University Management School, who presented on “Algorithms, Performativity and Governability.” Introna explored different “actor[s] in the dance of agency,” observing that the code of an algorithm is distinct from its “doing,” or performance. Lisa Gitelman, NYU professor of English and Media, Culture, and Communication, and Matthew Jones, a historian of science at Columbia University, offered stimulating responses.
Daniel Neyland of Goldsmiths, University of London, presented his work on “Bearing Accountable Witness to the Ethical Algorithm” in the final full panel. In it, Neyland details his rich experience exploring the ethics of designing a CCTV system. Mike Ananny, of the University of Southern California and Karrie Karahalios, of University of Illinois at Urbana-Champaign, responded with rich discussion of how we might frame ethical questions related to algorithms.
The conference concluded with Evgeny Morozov, author of To Save Everything, Click Here, and Paul Dourish of the University of California, Irvine, each offered concluding thoughts on the two-day event, noting common themes, additional implications, and alternative ways of framing or analyzing many of the issues discussed at the conference.
Audience members came from all over the United States and Europe to attend, forging interesting interdisciplinary connections. All written materials related to the Governing Algorithms conference are available on the conference web site, as is video of most all of the talks.
Posted on June 14, 2013