Human Computer Interaction, Human Centered AI, health communication, restorative justice
Assistant professor at UC, Berkeley School of Information
| Website | https://niloufar.org |
Defense teams face challenges assessing and contesting reliability due to difficulties (a) navigating how these tools are developed and used, (b) overcoming judges and jurors' non-critical perceptions of the technology, and (c) gathering technical expertise. Technical and policy interventions are urgently needed to help public defenders overcome these challenges.
(Beyond) Reasonable Doubt: Challenges that Public Defenders Face in Scrutinizing AI in Court https://arxiv.org/abs/2403.13004 #chi2024
Accountable use of AI systems in high-stakes settings relies on making systems contestable. In this paper we study efforts to contest AI systems in practice by studying how public defenders scrutinize AI in court. We present findings from interviews with 17 people in the U.S. public defense community to understand their perceptions of and experiences scrutinizing computational forensic software (CFS) -- automated decision systems that the government uses to convict and incarcerate, such as facial recognition, gunshot detection, and probabilistic genotyping tools. We find that our participants faced challenges assessing and contesting CFS reliability due to difficulties (a) navigating how CFS is developed and used, (b) overcoming judges and jurors' non-critical perceptions of CFS, and (c) gathering CFS expertise. To conclude, we provide recommendations that center the technical, social, and institutional context to better position interventions such as performance evaluations to support contestability in practice.
Colorado's star DNA analyst intentionally manipulated data and affected more than 650 cases.
The U.S. government increasingly uses difficult to scrutinize, ML-based decision systems to convict and incarcerate people, such as facial recognition, gunshot detection, and probabilistic genotyping tools. Our #chi2024 paper asks what challenges do people face contesting these systems in court?
So the whole rich people buying social media the second they don't like what is on it is foreshadowing what's going to happen to all the big models isn't it?
Many models produce outputs that are hard to verify for an end user.
🏆 Our new #emnlp2023 paper won an outstanding paper award for showing that a secondary quality estimation model can help users decide when to rely on the model output.
We ran a controlled experiment showing that a calibrated quality estimation model can make physicians twice better at correctly deciding when to rely on a translation model output.
New paper preprint! How do Reddit moderators deal with AI-generated content? @travislloyd tells you all about it here:
https://hci.social/@travislloyd/111500466847327538
(also, follow him on Mastodon)
*NEW PAPER* Since the launch of ChatGPT, online communities are reckoning with an influx of AI-generated content (AIGC). How are they dealing with it? To find out, I interviewed 15 community moderators from top subreddits to discuss how they and their communities are responding. /1
When we warn the real threat of AI is how it’s used against people in the present, not the fantasies that some day computers might think for themselves, this is exactly the kind of thing we’re talking about: health insurers using AI to deny care.
Many models produce outputs that are hard to verify for an end-user. Our new #emnlp2023 paper shows that providing a quality estimation model, can make a user better at deciding when to rely on the model.
📣 I'm recruiting PhD students for fall 2024 to join my new research lab at the University of Michigan CSE!
Interested in social computing systems that empower people? Consider applying!
Please spread the word 📣
The iSchool and Government & Politics Dept at UMD invite applications for a tenure-track assistant professor position. This position is open to applicants from any fields within information and political sciences that apply data science concepts and methods in their work. We seek candidates who address information science concepts and methodologies to study important social science problems in the broad field of government and politics.
For more information, see
https://ejobs.umd.edu/postings/114626
The College of Information Studies (INFO) at the University of Maryland, College Park and the Department of Government and Politics (GVPT) in the College of Behavioral and Social Sciences (BSOS) invite applications for a tenure-track position at the assistant professor level. This position is open to applicants from any fields within information and political sciences that apply data science concepts and methods in their work. INFO and GVPT particularly seek candidates whose scholarship and teaching interests address information science concepts and methodologies to study important social science problems in the broad field of government and politics. We encourage candidates to demonstrate how their research contributes toward social justice, equity, diversity, and inclusion in the creation, management, and accessibility of information.The faculty member will have a joint appointment at INFO (75%) and GVPT (25%), with a tenure home in INFO. The faculty member will be expected to teach courses in both INFO and GVPT, as well as to contribute to teaching and service to the newly formed undergraduate Major in Social Data Science, which is a INFO-BSOS joint program. We seek a candidate who will build upon our record of educating information and political scientists committed to leadership, service, research, and innovative practice and impact.The faculty member will be expected to have a high-impact research agenda and to seek external funding for research. The faculty member will also teach at the undergraduate and graduate levels, provide research advising to students at all levels, and engage in service to the profession and shared governance within the university.In particular candidates whose scholarship speaks to the following broad substantive topics are encouraged to apply:Social media politicsInformation disorder (misinformation, disinformation, etc.)Online political and civic participationPolicy and governance for emerging technologiesDigital democracy and data justiceData science for policy analysisRole of political institutions in production and dissemination of informationPolitical and moral values of information systemsThe political economy of information systemsWhile these areas of expertise are of particular interest, candidates with expertise and interest in other relevant topics that apply data science methods are encouraged to apply. Candidates whose work recognizes the information norms and assets of marginalized communities particularly relating to the topics above and other relevant topics, are especially encouraged to apply.Position Duties/Responsibilities:This is a tenure-track appointment. Tenure-track Assistant Professors are hired for 3 years, with potential renewal for three more years and tenure review. Salary and benefits are competitive based upon qualifications and experience. Tenure-track and tenured faculty typically have 9-month appointments, with opportunities for grant-funded research and university-funded teaching in summers. For this position, the successful candidate will be involved in:Developing and conducting high-impact research agenda and disseminating research resultsDesigning and developing innovative information studies curriculaCrafting exceptional educational experiences for studentsContributing to relevant professional communitiesParticipating in shared governance
Mor Naaman
Paris Marx
Farnaz Jahanbakhsh
jessica vitak (she/her)