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.

AI creates and exacerbates privacy risks, yet practitioners lack effective resources to identify and mitigate these risks. We present Privy, a tool that guides practitioners without privacy expertise through structured privacy impact assessments to: (i) identify relevant risks in novel AI product concepts, and (ii) propose appropriate mitigations. Privy was shaped by a formative study with 11 practitioners, which informed two versions — one LLM-powered, the other template-based. We evaluated these two versions of Privy through a between-subjects, controlled study with 24 separate practitioners, whose assessments were reviewed by 13 independent privacy experts. Results show that Privy helps practitioners produce privacy assessments that experts deemed high quality: practitioners identified relevant risks and proposed appropriate mitigation strategies. These effects were augmented in the LLM-powered version. Practitioners themselves rated Privy as being useful and usable, and their feedback illustrates how it helps overcome long-standing awareness, motivation, and ability barriers in privacy work.

In pseudonymous online fora like Reddit, the benefits of self-disclosure are often apparent to users (e.g., I can vent about my in-laws to understanding strangers), but the privacy risks are more abstract (e.g., will my partner be able to tell that this is me?). Prior work has sought to develop natural language processing (NLP) tools that help users identify potentially risky self-disclosures in their text, but none have been designed for or evaluated with the users they hope to protect. Absent this assessment, these tools will be limited by the social-technical gap: users need assistive tools that help them make informed decisions, not paternalistic tools that tell them to avoid self-disclosure altogether. To bridge this gap, we conducted a study with N =21 Reddit users; we had them use a state-of-the-art NLP disclosure detection model on two of their authored posts and asked them questions to understand if and how the model helped, where it fell short, and how it could be improved to help them make more informed decisions. Despite its imperfections, users responded positively to the model and highlighted its use as a tool that can help them catch mistakes, inform them of risks they were unaware of, and encourage self-reflection. However, our work also shows how, to be useful and usable, AI for supporting privacy decision-making must account for posting context, disclosure norms, and users' lived threat models, and provide explanations that help contextualize detected risks.

Yao Dou, Isadora Krsek, Tarek Naous, Anubha Kabra, Sauvik Das, Alan Ritter, Wei Xu. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

User adoption of security and privacy (S&P) best practices remains low, despite sustained efforts by researchers and practitioners. Social influence is a proven method for guiding user S&P behavior, though most work has focused on studying peer influence, which is only possible with a known social graph. In a study of 104 Facebook users, we instead demonstrate that crowdsourced S&P suggestions are significantly influential. We also tested how reflective writing affected participants’ S&P decisions, with and without suggestions. With reflective writing, participants were less likely to accept suggestions — both social and Facebook default suggestions. Of particular note, when reflective writing participants were shown the Facebook default suggestion, they not only rejected it but also (unknowingly) configured their settings in accordance with expert recommendations. Our work suggests that both non-personal social influence and reflective writing can positively influence users’ S&P decisions, but have negative interactions.