People candidly discuss sensitive topics online under the perceived safety of anonymity; yet, for many, this perceived safety is tenuous, as miscalibrated risk perceptions can lead to over-disclosure. Recent advances in Natural Language Processing (NLP) afford an unprecedented opportunity to present users with quantified disclosure-based re-identification risk (i.e., "population risk estimates", PREs). How can PREs be presented to users in a way that promotes informed decision-making, mitigating risk without encouraging unnecessary self-censorship? Using design fictions and comic-boarding, we story-boarded five design concepts for presenting PREs to users and evaluated them through an online survey with N = 44 Reddit users. We found participants had detailed conceptions of how PREs may impact risk awareness and motivation, but envisioned needing additional context and support to effectively interpret and act on risks. We distill our findings into four key design recommendations for how best to present users with quantified privacy risks to support informed disclosure decision-making.
As data privacy continues to be a crucial human-right concern as recognized by the UN, regulatory agencies have demanded developers obtain user permission before accessing user-sensitive data. Mainly through the use of privacy policies statements, developers fulfill their legal requirements to keep users abreast of the requests for their data. In addition, platforms such as Android enforces explicit permission request using the permission model. Nonetheless, recent research has shown that service providers hardly make full disclosure when requesting data in these statements. Neither is the cur…
Decoding the Digital Fine Print: Navigating the potholes in Terms of service/ use of GenAI tools against the emerging need for Transparent and Trustworthy Tech Futures
In social Virtual Reality (VR), users are embodied in avatars and interact with other users in a face-to-face manner using avatars as the medium. With the advent of social VR, people with disabilities (PWD) have shown an increasing presence on this new social media. With their unique disability identity, it is not clear how PWD perceive their avatars and whether and how they prefer to disclose their disability when presenting themselves in social VR. We fill this gap by exploring PWD's avatar perception and disability disclosure preferences in social VR. Our study involved two steps. We first …
Users disclose ever-increasing amounts of personal data on Social Network Service platforms (SNS). Unless SNSs' policies are privacy friendly, this leaves them vulnerable to privacy risks because they ignore the privacy policies. Designers and regulators have pushed for shorter, simpler and more prominent privacy policies, however the evidence that transparent policies increase informed consent is lacking. To answer this question, we conducted an online experiment with 214 regular Facebook users asked to join a fictitious SNS. We experimentally manipulated the privacy-friendliness of SNS's pol…
Carnegie Mellon University（卡内基梅隆大学）；University of Texas at Austin（德克萨斯大学奥斯汀分校）
In many real world contexts, successful human-AI collaboration requires humans to productively integrate complementary sources of information into AI-informed decisions. However, in practice human decision-makers often lack understanding of what information an AI model has access to in relation to themselves. There are few available guidelines regarding how to effectively communicate about unobservables: features that may influence the outcome, but which are unavailable to the model. In this work, we conducted an online experiment to understand whether and how explicitly communicating potentia…
Differential privacy (DP) is a mathematical privacy notion increasingly deployed across government and industry. With DP, privacy protections are probabilistic: they are bounded by the privacy budget parameter, $ε$. Prior work in health and computational science finds that people struggle to reason about probabilistic risks. Yet, communicating the implications of $ε$ to people contributing their data is vital to avoiding privacy theater -- presenting meaningless privacy protection as meaningful -- and empowering more informed data-sharing decisions. Drawing on best practices in risk communicat…
"Even GPT Can Reject Me": Conceptualizing Abrupt Refusal Secondary Harm (ARSH) and Reimagining Psychological AI Safety with Compassionate Completion Standard (CCS)
Symbiotic Future AI（共生未来人工智能）；Warner School of Education, University of Rochester（罗切斯特大学沃纳教育学院）
Systems for language-guided human-robot interaction must satisfy two key desiderata for broad adoption: adaptivity and learning efficiency. Unfortunately, existing instruction-following agents cannot adapt, lacking the ability to incorporate online natural language supervision, and even if they could, require hundreds of demonstrations to learn even simple policies. In this work, we address these problems by presenting Language-Informed Latent Actions with Corrections (LILAC), a framework for incorporating and adapting to natural language corrections - "to the right," or "no, towards the book"…
