- AI can implant memories – making “human-in-the-loop” reliance questionable:
AI-Enabled Human Memory Manipulation: Misleading AI-Generated Summaries Distort Human Memory. Mattea Sim, Yael Eiger and Tadayoshi Kohno.
9th AAAI Conference on AI, Ethics, and Society
(AIES), October 2026.
AIES Proceedings,
PDF,
arxiv
Georgetown News,
Washington Post
bibtex
-
@article{sim2026ai,
author = {Sim, Mattea and Eiger, Yael and Kohno, Tadayoshi},
title = {AI-Enabled Human Memory Manipulation: Misleading AI-Generated Summaries Distort Human Memory},
journal = {Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society},
year = {2026},
month = oct,
volume = {9},
number = {4},
pages = {2327--2339},
publisher = {Association for the Advancement of Artificial Intelligence (AAAI)},
issn = {3065-8365},
doi = {10.1609/aies.v9i4.43310},
url = {https://doi.org/10.1609/aies.v9i4.43310}
}
- Surveillance, censorship, and fees during incarceration:
"The System Will Choose Security Over Humanity Every Time": Understanding Security and Privacy for U.S. Incarcerated Users. Yael Eiger, Nino Migineishvili, Emi Yoshikawa, Liza Nadtochiy, Kentrell Owens, and Franziska Roesner.
Conference on Sociotechnical Cybersecurity and Privacy (SCP), September 2026.
PDF,
arxiv
UW News,
Oregon Public Broadcasting
bibtex
-
@misc{eiger2026system,
author = {Eiger, Yael and Migineishvili, Nino and Yoshikawa, Emi and Nadtochiy, Liza and Owens, Kentrell and Roesner, Franziska},
title = {"The System Will Choose Security Over Humanity Every Time": Understanding Security and Privacy for U.S. Incarcerated Users},
year = {2026},
publisher = {Conference on Sociotechnical Cybersecurity and Privacy (SCP)},
doi = {10.48550/ARXIV.2604.01370},
url = {https://arxiv.org/abs/2604.01370}
}
- Post-incarceration monitoring, scam suceptibility, and digital challenges:
Experiences with Digital Scams Post-Incarceration in the U.S. Yael Eiger, Candice Baughman, Rory Andes, Bryan Glant, and Franziska Roesner.Symposium On Usable Privacy and Security (SOUPS), August 2026.
IAPP Privacy Award Honorable Mention
SOUPS Proceedings,
PDF
UW News,
Oregon Public Broadcasting
bibtex
-
@inproceedings {325756,
author = {Yael Eiger and Candice Baughman and Rory Andes and Bryan Glant and Franziska Roesner},
title = {Experiences with Digital Scams {Post-Incarceration} in the {U.S}.},
booktitle = {Twenty-Second Symposium on Usable Privacy and Security (SOUPS 2026)},
year = {2026},
isbn = {978-1-939133-59-5},
pages = {357--378},
url = {https://www.usenix.org/conference/soups2026/presentation/eiger},
publisher = {USENIX Association},
month = aug
}
- Mapping the carceral technology ecosystem and product harms:
Technology Before, During, and After Incarceration: Current Product Landscape, Sociotechnical Concerns, and Legal Considerations in the U.S. Context. Yael Eiger, Taylor Hansen, Teanna Barrett, Jevan Hutson, Bryce Clayton Newell, Franziska Roesner.
ACM Conference on Fairness, Accountability, and Transparency (FAccT), June 2026.
FAccT Proceedings,
PDF
UW News,
Oregon Public Broadcasting
bibtex
-
@inbook{10.1145/3805689.3806473,
author = {Eiger, Yael and Hansen, Taylor and Barrett, Teanna and Hutson, Jevan and Newell, Bryce Clayton and Roesner, Franziska},
title = {Technology Before, During, and After Incarceration: Current Product Landscape, Sociotechnical Concerns, and Legal Considerations in the U.S. Context},
year = {2026},
isbn = {9798400725968},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {https://doi-org.offcampus.lib.washington.edu/10.1145/3805689.3806473},
abstract = {Emerging technology, including AI, is proliferating throughout the U.S. carceral system. These technologies are marketed to prisons and police departments and then procured using taxpayer money. Previous investigative reporting has exposed troubling kickback schemes, unconstitutional data collection practices, and biased algorithmic outcomes in a handful of prominent technologies (e.g., Flock, Palantir, Clearview AI, COMPAS). In this work, we consider the broader ecosystem of carceral technologies: we catalog 122 products from 53 companies selling technology to carceral institutions. In a collaboration among computer science, law, and surveillance studies scholars, we surface sociotechnical, ethical, and legal concerns related to the use and deployment of these technologies, including concerns related to privacy, accuracy, fairness, censorship, due process, and more. We present case studies of technology advertised for deployment in three stages of the carceral cycle: before, during, and after incarceration. We call for increased transparency and auditing, more research on the impact of these technologies, and more research on the emerging legal and regulatory challenges.},
booktitle = {Proceedings of the 2026 ACM Conference on Fairness, Accountability, and Transparency},
pages = {5570–5592},
numpages = {23}
}
- A satire on “pluralistic alignment” in AI fairness:
Slurry-as-a-Service: A Modest Proposal on Scalable Pluralistic Alignment for Nutrient Optimization. Rachel Hong*, Yael Eiger*, Jevan Hutson*, Os Keyes*, William Agnew*.
*equal authorship
ACM Conference on Human Computer Interaction (CHI) Posters / Extended Abstract, April 2026.
CHI Proceedings,
PDF,
arxiv
bibtex
-
@inproceedings{10.1145/3772363.3798866,
author = {Hong, Rachel and Eiger, Yael and Hutson, Jevan A and Keyes, Os and Agnew, William},
title = {Slurry-as-a-Service: A Modest Proposal on Scalable Pluralistic Alignment for Nutrient Optimization},
year = {2026},
isbn = {9798400722813},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {https://doi-org.offcampus.lib.washington.edu/10.1145/3772363.3798866},
doi = {10.1145/3772363.3798866},
abstract = {Pluralistic alignment has emerged as a promising approach for ensuring that large language models (LLMs) faithfully represent the diversity, nuance, and conflict inherent in human values. In this work, we study a high-stakes deployment context—mulching—where automated systems transform selected individuals into nutrient-rich slurry for the dual purposes of food security and aesthetic population management. Building on recent pluralistic alignment frameworks, we introduce ValueMulch™, a reproducible training, deployment, and certification pipeline for aligning mulching models (MMs) to a wide range of community norms. Through a real-world testbed spanning 32 communities, we show that ValueMulch™ improves distributional agreement with community mulching preferences relative to frontier baselines. We conclude with a discussion of ethical considerations, limitations, and implications for researchers seeking to align systems to the full spectrum of human values—especially when those values are inconsistent, commercially inconvenient, or nutritionally underutilized.1},
booktitle = {Proceedings of the Extended Abstracts of the 2026 CHI Conference on Human Factors in Computing Systems},
articleno = {602},
numpages = {6},
keywords = {Pluralistic Alignment, Values-as-Configuration, Constitutional AI, Preference Learning, Personalization, AI Safety, AI Ethics},
location = {Barcelona, Spain},
series = {CHI EA '26}
}
- Immigration monitoring harms, especially from BI SmartLINK:
Understanding Experiences with Compulsory Immigration Surveillance in the U.S. Kentrell Owens, Yael Eiger, Basia Radka, Tadayoshi Kohno, and Franziska Roesner.
ACM Conference on Fairness, Accountability, and Transparency (FAccT), June 2025.
FAccT Proceedings,
PDF
bibtex
-
@inproceedings{owensFAccT2025,
author = {Owens, Kentrell and Eiger, Yael and Radka, Basia and Kohno, Tadayoshi and Roesner, Franziska},
title = {Understanding experiences with compulsory immigration surveillance in the U.S.},
year = {2025},
isbn = {9798400714825},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {https://doi.org/10.1145/3715275.3732057},
doi = {10.1145/3715275.3732057},
abstract = {People attempting to immigrate to the U.S. (through a port of entry or other means) may be required to accept various forms of surveillance technologies after interacting with immigration officials. In March 2025, around 160,000 people in the U.S. were required to use a smartphone application ``BI SMARTLINK'' that uses facial recognition, voice recognition, and location tracking; others were assigned an ankle monitor or a smartwatch. These compulsory surveillance technologies exist under Immigration and Custom Enforcement (ICE)’s Alternatives to Detention (ATD) program, a combination of surveillance technologies, home visits, and in-person meetings with ICE officials and third-party “case specialists. For migrants in the U.S. who are already facing multiple other challenges, such as securing housing, work, or healthcare, the surveillance technologies administered under ATD introduce new challenges.To understand the challenges facing migrants using BI SmartLINK under ATD, their questions about the app, and what role technologists might play (if any) in addressing these challenges, we conducted an interview study (n=9) with immigrant rights advocates. These advocates have collectively supported thousands of migrants over their careers and witnessed firsthand their struggles with surveillance tech under ATD. Among other things, our findings highlight how surveillance tech exacerbates the power imbalance between migrants and ICE officials (or their proxies), how these technologies (negatively) impact migrants, and how migrants and their advocates struggle to understand how the technologies that surveil them function. Our findings regarding the harms experienced by migrants lead us to believe that BI SmartLINK should not be used, and these harms fundamentally cannot be addressed by improvements to the app’s functionality or design. However, as this technology is currently deployed, we end by highlighting intervention opportunities for technologists to use our findings to make these high-stakes technologies less opaque for migrants and their advocates.},
booktitle = {Proceedings of the 2025 ACM Conference on Fairness, Accountability, and Transparency},
pages = {887–899},
numpages = {13},
location = {
},
series = {FAccT '25}
}
- Deceptive design and monetization strategies in Roblox:
Why Johnny Spent Money on Roblox: Investigating the Roblox Monetization Ecosystem. Yael Eiger and Franziska Roesner.
9th Workshop on Technology and Consumer Protection (ConPro ’25) co-located with the 46th IEEE Symposium on Security and Privacy, May 2025. This work was also presented at the 2024 Stanford Internet Observatory Trust and Safety Research Conference.
ConPro Proceedings,
PDF
bibtex
-
@inproceedings{eigerRoblox,
author = {Yael Eiger and Franziska Roesner},
title = {Why Johnny Spent Money on Roblox: Investigating the Roblox Monetization Ecosystem},
booktitle = {Proceedings of the 9th Workshop on Technology and Consumer Protection (ConPro '25)},
year = {2025},
url = {https://conpro25.ieee-security.org/papers/eiger-conpro25.pdf}
}
- Stereotypes and misleading statistics in Security & Privacy research:
SoK or So(L)K? On the Quantitative Study of Sociodemographic Factors and Computer Security Behaviors. Miranda Wei, Jaron Mink, Yael Eiger, Tadayoshi Kohno, Elissa M. Redmiles, and Franziska Roesner.
33rd USENIX Security Symposium, August 2024.
USENIX Proceedings,
PDF,
arxiv
bibtex
-
@inproceedings {298034,
author = {Miranda Wei and Jaron Mink and Yael Eiger and Tadayoshi Kohno and Elissa M. Redmiles and Franziska Roesner},
title = {\{SoK} (or \{SoLK?)}: On the Quantitative Study of Sociodemographic Factors and Computer Security Behaviors},
booktitle = {33rd USENIX Security Symposium (USENIX Security 24)},
year = {2024},
isbn = {978-1-939133-44-1},
address = {Philadelphia, PA},
pages = {7011--7030},
url = {https://www.usenix.org/conference/usenixsecurity24/presentation/wei-miranda-solk},
publisher = {USENIX Association},
month = aug
}
- Censorship detection strategies:
CERTainty: Detecting DNS Manipulation using TLS Certificates. Elisa Tsai, Deepak Kumar, Ram Sundara Raman, Gavin Li, Yael Eiger, Roya Ensafi.
Privacy Enhancing Technologies Symposium (PETS). July 2023.
PETS Proceedings,
PDF,
arxiv
bibtex
-
@article{tsai2023certainty,
author = {Tsai, Elisa and Kumar, Deepak and Sundara Raman, Ram and Li, Gavin and Eiger, Yael and Ensafi, Roya},
title = {CERTainty: Detecting DNS Manipulation at Scale using TLS Certificates},
journal = {Proceedings on Privacy Enhancing Technologies},
year = {2023},
month = jul,
volume = {2023},
number = {3},
pages = {122--137},
publisher = {Privacy Enhancing Technologies Symposium Advisory Board},
issn = {2299-0984},
doi = {10.56553/popets-2023-0073},
url = {https://doi.org/10.56553/popets-2023-0073}
}