Over the past few years, the two dominant app platforms made major improvements to their policies surrounding child-directed apps. While prior work repeatedly demonstrated that privacy issues were prevalent in child-directed apps, it is unclear whether platform policies can lead child-directed apps to comply with privacy requirements, when laws alone have not. To understand the effect of recent changes in platform policies (e.g., whether they result in greater levels of compliance with applicable privacy laws), we conducted a large-scale measurement study of the privacy behaviors of 7,377 child-directed Android apps, as well as a follow-up survey with some of their developers. We observed a drastic decrease in the number of apps that transmitted personal data without verifiable parental consent and an increase in the number of apps that encrypted their transmissions using TLS. However, improper use of third-party SDKs still led to privacy issues (e.g., inaccurate disclosures in apps’ privacy labels). Our analysis of apps’ privacy practices over a period of a few months in 2023 and a comparison of our results with those observed a few years ago demonstrate gradual improvements in apps’ privacy practices over time. We discuss how app platforms can further improve their policies and emphasize the role of enforcement in making such policies effective.

The Android virtualization technique allows an app to create independent virtual environments running on top of the Android native one, where multiple apps can be executed simultaneously. While the technique has legitimate uses, attackers have identified ways to exploit it. According to the state-of-art, virtualization-based malware is a significant threat: researchers have found 71,303 malicious samples. Defence mechanisms have already been developed to find virtualization-based malware and to detect or prevent virtualization-based repackaging attacks.In this paper, we offer three key contributions. First, we experimentally evaluate the existing defence mechanisms by identifying their limitations and demonstrating how they can be bypassed. Second, we design and develop a new defence mechanism, called Matrioska, that overcomes the limitations of the state-of-art by detecting the intrinsic features of the virtualization technique. Third, we evaluate the effectiveness of Matrioska with respect to the state-of-art against two datasets of apps. Overall, Matrioska achieves a higher accuracy (99% vs. 71%) when searching for virtualization usage and a lower false positive (10 vs. 23) and false negative rate (14 vs. 39) when detecting a virtualization-based repackaging attack.

Like most modern software, secure messaging apps rely on thirdparty components to implement important app functionality. Although this practice reduces engineering costs, it also introduces the risk of inadvertent privacy breaches due to misconfiguration errors or incomplete documentation. Our research investigated secure messaging apps' usage of Google's Firebase Cloud Messaging (FCM) service to send push notifications to Android devices. We analyzed 21 popular secure messaging apps from the Google Play Store to determine what personal information these apps leak in the payload of push notifications sent via FCM. Of these apps, 11 leaked metadata, including user identifiers (10 apps), sender or recipient names (7 apps), and phone numbers (2 apps), while 4 apps leaked the actual message content. Furthermore, none of the data we observed being leaked to FCM was specifically disclosed in those apps' privacy disclosures. We also found several apps employing strategies to mitigate this privacy leakage to FCM, with varying levels of success. Of the strategies we identified, none appeared to be common, shared, or well-supported. We argue that this is fundamentally an economics problem: incentives need to be correctly aligned to motivate platforms and SDK providers to make their systems secure and private by default.

Internet users often neglect important security actions (e.g., installing security updates or changing passwords) because they interrupt users’ main task at inopportune times. Commitment devices, such as reminders and promises, have been found to be effective at reducing procrastination in other domains. In a series of online experiments ( \(n{\gt}3{,}000\) ), we explored the effects of reminders and promises on users’ willingness to change a compromised password. We find that adding an option to delay the task increases the share of people willing to eventually change their password considerably. Critically, the option to delay yields this overall increase without reducing the share of people choosing to change their password immediately. Additionally, most participants who promised to change their password later, or asked to be reminded to do so, indeed followed through on their commitment, leading to a net positive effect. Reminding participants of their previous commitment further increased this effect.

The network communication between Internet of Things (IoT) devices on the same local network has significant implications for platform and device interoperability, security, privacy, and correctness. Yet, the analysis of local home Wi-Fi network traffic and its associated security and privacy threats have been largely ignored by prior literature, which typically focuses on studying the communication between IoT devices and cloud end-points, or detecting vulnerable IoT devices exposed to the Internet. In this paper, we present a comprehensive and empirical measurement study to shed light on the local communication within a smart home deployment and its threats. We use a unique combination of passive network traffic captures, protocol honeypots, dynamic mobile app analysis, and crowdsourced IoT data from participants to identify and analyze a wide range of device activities on the local network. We then analyze these datasets to characterize local network protocols, security and privacy threats associated with them. Our analysis reveals vulnerable devices, insecure use of network protocols, and sensitive data exposure by IoT devices. We provide evidence of how this information is exfiltrated to remote servers by mobile apps and third-party SDKs, potentially for household fingerprinting, surveillance and cross-device tracking. We make our datasets and analysis publicly available to support further research in this area.

The California Consumer Privacy Act (CCPA) provides California residents with a range of enhanced privacy protections and rights. Our research investigated the extent to which Android app developers comply with the provisions of the CCPA that require them to provide consumers with accurate privacy notices and respond to "verifiable consumer requests" (VCRs) by disclosing personal information that they have collected, used, or shared about consumers for a business or commercial purpose. We compared the actual network traffic of 109 apps that we believe must comply with the CCPA to the data that apps state they collect in their privacy policies and the data contained in responses to "right to know" requests that we submitted to the app's developers. Of the 69 app developers who substantively replied to our requests, all but one provided specific pieces of personal data (as opposed to only categorical information). However, a significant percentage of apps collected information that was not disclosed, including identifiers (55 apps, 80%), geolocation data (21 apps, 30%), and sensory data (18 apps, 26%) among other categories. We discuss improvements to the CCPA that could help app developers comply with "right to know" requests and other related regulations.

The sharing of information between older adults and their friends, families, caregivers, and doctors promotes a collaborative approach to managing their emotional, mental, and physical well-being and health, prolonging independent living, and improving care quality and quality of life in general. However, information flow in collaborative systems is complex, not always transparent to elderly users, and may raise privacy and security concerns. Because older adults’ decisions about whether to engage in information exchange affect interpersonal communications and delivery of care, it is important to understand the factors and context that influence those decisions. Our work contributes empirical evidence and suggests a systematic approach. In this article, we present the results of semi-structured interviews with 46 older adults aged 65+ about their views on information collection, transmission, and sharing. We develop a detailed model of the contextual factors that combine in complex ways to affect older adults’ decision making about information sharing. We discuss how our comprehensive model compares to existing frameworks for analyzing information-sharing expectations and preferences. Finally, we suggest directions for future research and describe the practical implications of our model for the design and evaluation of collaborative information-sharing systems, as well as for policy and consumer protection.

As intelligent voice assistants become more widespread and the scope of their listening increases, they become attractive targets for attackers. In the future, a malicious actor could train voice assistants to listen to audio outside their purview, creating a threat to users' privacy and security. How can this misbehavior be detected? Due to the ambiguities of natural language, people may need to work in conjunction with algorithms to determine whether a given conversation should be heard. To investigate how accurately humans can perform this task, we developed a framework for people to conduct "Test Drives" of always-listening services: after submitting sample conversations, users receive instant feedback about whether these would have been captured. Leveraging a Wizard of Oz interface, we conducted a study with 200 participants to determine whether they could detect one of four types of attacks on three different services. We studied the behavior of individuals, as well as groups working collaboratively, and investigated the effects of task framing on performance. We found that individuals were able to successfully detect malicious apps at varying rates (7.5% to 75%), depending on the type of malicious attack, and that groups were highly successful when considered collectively. Our results suggest that the Test Drive framework can be an effective tool for studying user behaviors and concerns, as well as a potentially welcome addition to voice assistant app stores, where it could decrease privacy concerns surrounding always-listening services.

We investigate the privacy compliance processes followed by developers of child-directed mobile apps. While children’s online privacy laws have existed for decades in the US, prior research found relatively low rates of compliance. Yet, little is known about how compliance issues come to exist and how compliance processes can be improved to address them. Our results, based on surveys (n = 127) and interviews (n = 27), suggest that most developers rely on app markets to identify privacy issues, they lack complete understandings of the third-party SDKs they integrate, and they find it challenging to ensure that these SDKs are kept upto-date and privacy-related options are configured correctly. As a result, we find that well-resourced app developers outsource most compliance decisions to auditing services, and that smaller developers follow “best-effort” models, by assuming that their apps are compliant so long as they have not been rejected by app markets. We highlight the need for usable tools that help developers identify and fix mobile app privacy issues.

Abstract Privacy expectations during disasters differ significantly from nonemergency situations. This paper explores the actual privacy practices of popular disaster apps, highlighting location information flows. Our empirical study compares content analysis of privacy policies and government agency policies, structured by the contextual integrity framework, with static and dynamic app analysis documenting the personal data sent by 15 apps. We identify substantive gaps between regulation and guidance, privacy policies, and information flows, resulting from ambiguities and exploitation of exemptions. Results also indicate gaps between governance and practice, including the following: (a) Many apps ignore self‐defined policies; (b) while some policies state they “might” access location data under certain conditions, those conditions are not met as 12 apps included in our study capture location immediately upon initial launch under default settings; and (c) not all third‐party data recipients are identified in policy, including instances that violate expectations of trusted third parties.

Abstract It is commonly assumed that “free” mobile apps come at the cost of consumer privacy and that paying for apps could offer consumers protection from behavioral advertising and long-term tracking. This work empirically evaluates the validity of this assumption by comparing the privacy practices of free apps and their paid premium versions, while also gauging consumer expectations surrounding free and paid apps. We use both static and dynamic analysis to examine 5,877 pairs of free Android apps and their paid counterparts for differences in data collection practices and privacy policies between pairs. To understand user expectations for paid apps, we conducted a 998-participant online survey and found that consumers expect paid apps to have better security and privacy behaviors. However, there is no clear evidence that paying for an app will actually guarantee protection from extensive data collection in practice. Given that the free version had at least one thirdparty library or dangerous permission, respectively, we discovered that 45% of the paid versions reused all of the same third-party libraries as their free versions, and 74% of the paid versions had all of the dangerous permissions held by the free app. Likewise, our dynamic analysis revealed that 32% of the paid apps exhibit all of the same data collection and transmission behaviors as their free counterparts. Finally, we found that 40% of apps did not have a privacy policy link in the Google Play Store and that only 3.7% of the pairs that did reflected differences between the free and paid versions.

Commitment devices are a technique from behavioral economics that have been shown to mitigate the effects of present bias---the tendency to discount future risks and gains in favor of immediate gratifications. In this paper, we explore the feasibility of using commitment devices to nudge users towards complying with varying online security mitigations. Using two online experiments, with over 1,000 participants total, we offered participants the option to be reminded or to schedule security tasks in the future. We find that both reminders and commitment nudges can increase users' intentions to install security updates and enable two-factor authentication, but not to configure automatic backups. Using qualitative data, we gain insights into the reasons for postponement and how to improve future nudges. We posit that current nudges may not live up to their full potential, as the timing options offered to users may be too rigid.

Intelligent voice assistants (IVAs) and other voice-enabled devices already form an integral component of the Internet of Things and will continue to grow in popularity. As their capabilities evolve, they will move beyond relying on the wake-words today's IVAs use, engaging instead in continuous listening. Though potentially useful, the continuous recording and analysis of speech can pose a serious threat to individuals' privacy. Ideally, users would be able to limit or control the types of information such devices have access to. But existing technical approaches are insufficient for enforcing any such restrictions. To begin formulating a solution, we develop a systematic methodology for studying continuous-listening applications and survey architectural approaches to designing a system that enhances privacy while preserving the benefits of always-listening assistants.

Abstract As devices with always-on microphones located in people’s homes, smart speakers have significant privacy implications. We surveyed smart speaker owners about their beliefs, attitudes, and concerns about the recordings that are made and shared by their devices. To ground participants’ responses in concrete interactions, rather than collecting their opinions abstractly, we framed our survey around randomly selected recordings of saved interactions with their devices. We surveyed 116 owners of Amazon and Google smart speakers and found that almost half did not know that their recordings were being permanently stored and that they could review them; only a quarter reported reviewing interactions, and very few had ever deleted any. While participants did not consider their own recordings especially sensitive, they were more protective of others’ recordings (such as children and guests) and were strongly opposed to use of their data by third parties or for advertising. They also considered permanent retention, the status quo, unsatisfactory. Based on our findings, we make recommendations for more agreeable data retention policies and future privacy controls.

Web browser warnings should help protect people from malware, phishing, and network attacks. Adhering to warnings keeps people safer online. Recent improvements in warning design have raised adherence rates, but they could still be higher. And prior work suggests many people still do not understand them. Thus, two challenges remain: increasing both comprehension and adherence rates. To dig deeper into user decision making and comprehension of warnings, we performed an experience sampling study of web browser security warnings, which involved surveying over 6,000 Chrome and Firefox users in situ to gather reasons for adhering or not to real warnings. We find these reasons are many and vary with context. Contrary to older prior work, we do not find a single dominant failure in modern warning design---like habituation---that prevents effective decisions. We conclude that further improvements to warnings will require solving a range of smaller contextual misunderstandings.

Modern mobile operating systems implement an ask-on-first-use policy to regulate applications' access to private user data: the user is prompted to allow or deny access to a sensitive resource the first time an app attempts to use it. Prior research shows that this model may not adequately capture user privacy preferences because subsequent requests may occur under varying contexts. To address this shortcoming, we implemented a novel privacy management system in Android, in which we use contextual signals to build a classifier that predicts user privacy preferences under various scenarios. We performed a 37-person field study to evaluate this new permission model under normal device usage. From our exit interviews and collection of over 5 million data points from participants, we show that this new permission model reduces the error rate by 75% (i.e., fewer privacy violations), while preserving usability. We offer guidelines for how platforms can better support user privacy decision making.

Abstract We present a scalable dynamic analysis framework that allows for the automatic evaluation of the privacy behaviors of Android apps. We use our system to analyze mobile apps’ compliance with the Children’s Online Privacy Protection Act (COPPA), one of the few stringent privacy laws in the U.S. Based on our automated analysis of 5,855 of the most popular free children’s apps, we found that a majority are potentially in violation of COPPA, mainly due to their use of thirdparty SDKs. While many of these SDKs offer configuration options to respect COPPA by disabling tracking and behavioral advertising, our data suggest that a majority of apps either do not make use of these options or incorrectly propagate them across mediation SDKs. Worse, we observed that 19% of children’s apps collect identifiers or other personally identifiable information (PII) via SDKs whose terms of service outright prohibit their use in child-directed apps. Finally, we show that efforts by Google to limit tracking through the use of a resettable advertising ID have had little success: of the 3,454 apps that share the resettable ID with advertisers, 66% transmit other, non-resettable, persistent identifiers as well, negating any intended privacy-preserving properties of the advertising ID.

Text passwords---a frequent vector for account compromise, yet still ubiquitous---have been studied for decades by researchers attempting to determine how to coerce users to create passwords that are hard for attackers to guess but still easy for users to type and memorize. Most studies examine one password or a small number of passwords per user, and studies often rely on passwords created solely for the purpose of the study or on passwords protecting low-value accounts. These limitations severely constrain our understanding of password security in practice, including the extent and nature of password reuse, password behaviors specific to categories of accounts (e.g., financial websites), and the effect of password managers and other privacy tools.

Current smartphone operating systems regulate application permissions by prompting users on an ask-on-first-use basis. Prior research has shown that this method is ineffective because it fails to account for context: the circumstances under which an application first requests access to data may be vastly different than the circumstances under which it subsequently requests access. We performed a longitudinal 131-person field study to analyze the contextuality behind user privacy decisions to regulate access to sensitive resources. We built a classifier to make privacy decisions on the user's behalf by detecting when context has changed and, when necessary, inferring privacy preferences based on the user's past decisions and behavior. Our goal is to automatically grant appropriate resource requests without further user intervention, deny inappropriate requests, and only prompt the user when the system is uncertain of the user's preferences. We show that our approach can accurately predict users' privacy decisions 96.8% of the time, which is a four-fold reduction in error rate compared to current systems.

Abstract Although Tor has state-of-the art anticensorship measures, users in heavily censored environments will not be able to connect to Tor if they cannot configure their connections. We perform the first usability evaluation of Tor Launcher, the graphical user interface (GUI) that Tor Browser uses to configure connections to Tor. Our study shows that 79% (363 of 458) of user attempts to connect to Tor in simulated censored environments failed. We found that users were often frustrated during the process and tried options at random. In this paper, we measure potential usability issues, discuss design constraints unique to Tor, and provide recommendations based on what we learned to help more users connect to Tor while reducing the time they take to do so. Tor Browser incorporated the changes proposed by this study.

The Security Behavior Intentions Scale (SeBIS) measures the computer security attitudes of end-users. Because intentions are a prerequisite for planned behavior, the scale could therefore be useful for predicting users' computer security behaviors. We performed three experiments to identify correlations between each of SeBIS's four sub-scales and relevant computer security behaviors. We found that testing high on the awareness sub-scale correlated with correctly identifying a phishing website; testing high on the passwords sub-scale correlated with creating passwords that could not be quickly cracked; testing high on the updating sub-scale correlated with applying software updates; and testing high on the securement sub-scale correlated with smartphone lock screen usage (e.g., PINs). Our results indicate that SeBIS predicts certain computer security behaviors and that it is a reliable and valid tool that should be used in future research.

To prevent unauthorized parties from accessing data stored on their smartphones, users have the option of enabling a "lock screen" that requires a secret code (e.g., PIN, drawing a pattern, or biometric) to gain access to their devices. We present a detailed analysis of the smartphone locking mechanisms currently available to billions of smartphone users worldwide. Through a month-long field study, we logged events from a panel of users with instrumented smartphones (N=134). We are able to show how existing lock screen mechanisms provide users with distinct tradeoffs between usability (unlocking speed vs. unlocking frequency) and security. We find that PIN users take longer to enter their codes, but commit fewer errors than pattern users, who unlock more frequently and are very prone to errors. Overall, PIN and pattern users spent the same amount of time unlocking their devices on average. Additionally, unlock performance seemed unaffected for users enabling the stealth mode for patterns. Based on our results, we identify areas where device locking mechanisms can be improved to result in fewer human errors -- increasing usability -- while also maintaining security.

We present the results of an online survey of smartphone unlocking (N=8,286) that we conducted in eight different countries. The goal was to investigate differences in attitudes towards smartphone unlocking between different national cultures. Our results show that there are indeed significant differences across a range of categories. For instance, participants in Japan considered the data on their smartphones to be much more sensitive than those in other countries, and respondents in Germany were 4.5 times more likely than others to say that protecting data on their smartphones was important. The results of this study shed light on how motivations to use various security mechanisms are likely to differ from country to country.

A basic understanding of online privacy is essential to being an informed digital citizen, and therefore basic privacy education is becoming ever more necessary. Recently released high school and college computer science curricula acknowledge the significantly increased importance of fundamental knowledge about privacy, but do not yet provide concrete content in the area. To address this need, over the past two years, we have developed the Teaching Privacy Project (TPP) curriculum, http://teachingprivacy.org, which educates the general public about online privacy issues. We performed a pilot of our curriculum in a university course for non-CS majors and found that it was effective: weeks after last being exposed, students' privacy attitudes had shifted. In this paper, we describe our curriculum, our evaluation of it in the classroom, and our vision for future privacy education.

We are approaching an environment where ubiquitous computing devices will constantly accept input via audio and video channels: kiosks that determine demographic information of passersby, gesture controlled home entertainment systems and audio controlled wearable devices are just a few examples. To enforce the principle of least privilege, recent proposals have suggested technical approaches to limit third-party applications to receiving only the data they need, rather than entire audio or video streams. For users to make informed privacy decisions, applications will still need to communicate what data they are accessing and indicators will be needed to communicate this information. We performed several crowdsourcing experiments to examine how potential users might conceptualize and understand privacy indicators on ubiquitous sensing platforms.

Despite the plethora of security advice and online education materials offered to end-users, there exists no standard measurement tool for end-user security behaviors. We present the creation of such a tool. We surveyed the most common computer security advice that experts offer to end-users in order to construct a set of Likert scale questions to probe the extent to which respondents claim to follow this advice. Using these questions, we iteratively surveyed a pool of 3,619 computer users to refine our question set such that each question was applicable to a large percentage of the population, exhibited adequate variance between respondents, and had high reliability (i.e., desirable psychometric properties). After performing both exploratory and confirmatory factor analysis, we identified a 16-item scale consisting of four sub-scales that measures attitudes towards choosing passwords, device securement, staying up-to-date, and proactive awareness.

Most laptops and personal computers have webcams with LED indicators to notify users when they are recording. Because hackers use surreptitiously captured webcam recordings to extort users, we explored the effectiveness of these indicators under varying circumstances and how they could be improved. We observed that, on average, fewer than half of our participants (45%) noticed the existing indicator during computer-based tasks. When seated in front of the computer performing a paper-based task, only 5% noticed the indicator. We performed a followup experiment to evaluate a new indicator and observed that adding onscreen glyphs had a significant impact on both computer-based and non-computer-based tasks (93% and 59% noticed the new indicator, respectively). We discuss how our results can be integrated into current systems, as well as future ubiquitous computing systems.

While individual differences in decision-making have been examined within the social sciences for several decades, they have only recently begun to be applied by computer scientists to examine privacy and security attitudes (and ultimately behaviors). Specifically, several researchers have shown how different online privacy decisions are correlated with the "Big Five" personality traits. In this paper, we show that the five factor model is actually a weak predictor of privacy attitudes, and that other well-studied individual differences in the psychology literature are much stronger predictors. Based on this result, we introduce the new paradigm of psychographic targeting of privacy and security mitigations: we believe that the next frontier in privacy and security research will be to tailor mitigations to users' individual differences. We explore the extensive work on choice architecture and "nudges," and discuss the possible ways it could be leveraged to improve security outcomes by personalizing privacy and security mitigations to specific user traits.

Although frequent stories in the popular media have raised awareness about online privacy, most young people do not have a very good handle on what the specific issues are, nor the practical steps they can take to manage them. Teachers recognize that all their students' from future engineers to those totally bored by science need a realistic understanding of how online privacy works, so they can protect themselves online. In fact, the latest CS curricular recommendations include privacy but there is no comprehensive set of field-tested teaching materials. To address this, we are developing TROPE (Teachers' Resources for Online Privacy Education), a set of classroom-ready teaching materials (teachingprivacy.org). TROPE will provide educators with lesson modules, interactive demonstrations, and a teachers' guide, so they can readily integrate privacy into high school and college classes. Our goal for this workshop is twofold. First, we will introduce educators to TROPE and provide guidance on how they can cover privacy-related topics in their classrooms without being subject-matter experts. Second, we will solicit feedback and on-the-ground stories; by gaining a better understanding of specific problems faced by educators and students, we can increase TROPE's utility to teachers. We will provide teachers with up-to-date technical information about online privacy, including relevant highlights from our research; hands-on activities illustrating principles of online privacy; and an overview of the materials we are creating for TROPE. This will be an interactive workshop, driven by participants' questions, experiences, and interests. For CS educators at all levels; laptop or tablet recommended.

In addition to storing a plethora of sensitive personal and work information, smartphones also store sensor data about users and their daily activities. In order to understand users' behaviors and attitudes towards the security of their smartphone data, we conducted 28 qualitative interviews. We examined why users choose (or choose not) to employ locking mechanisms (e.g., PINs) and their perceptions and awareness about the sensitivity of the data stored on their devices. We performed two additional online experiments to quantify our interview results and the extent to which sensitive data could be found in a user's smartphone-accessible email archive. We observed a strong correlation between use of security features and risk perceptions, which indicates rational behavior. However, we also observed that most users likely underestimate the extent to which data stored on their smartphones pervades their identities, online and offline.

In Apple's iOS 6, when an app requires access to a protected resource (e.g., location or photos), the user is prompted with a permission request that she can allow or deny. These permission request dialogs include space for developers to optionally include strings of text to explain to the user why access to the resource is needed. We examine how app developers are using this mechanism and the effect that it has on user behavior. Through an online survey of 772 smartphone users, we show that permission requests that include explanations are significantly more likely to be approved. At the same time, our analysis of 4,400 iOS apps shows that the adoption rate of this feature by developers is relatively small: around 19% of permission requests include developer-specified explanations. Finally, we surveyed 30 iOS developers to better understand why they do or do not use this feature.

We performed a laboratory experiment to study the privacy tradeoff offered by Facebook Connect: disclosing Facebook profile data to third-party websites for the convenience of logging in without creating separate accounts. We controlled for trustworthiness and amount of information each website requested, as well as the consent dialog layout. We discovered that these factors had no observable effects, likely because participants did not read the dialogs. Yet, 15% still refused to use Facebook Connect, citing privacy concerns. A likely explanation for subjects ignoring the dialogs while also understanding the privacy tradeoff - our exit survey indicated that 88% broadly understood what data would be collected - is that subjects were already familiar with the dialogs prior to the experiment. We discuss how our results demonstrate informed consent, but also how habituation prevented subjects from understanding the nuances between individual websites' data collection policies.

Password meters tell users whether their passwords are "weak" or "strong." We performed a laboratory experiment to examine whether these meters influenced users' password selections when they were forced to change their real passwords, and when they were not told that their passwords were the subject of a study. We observed that the presence of meters yielded significantly stronger passwords. We performed a followup field experiment to test a different scenario: creating a password for an unimportant account. In this scenario, we found that the meters made no observable difference: participants simply reused weak passwords that they used to protect similar low-risk accounts. We conclude that meters result in stronger passwords when users are forced to change existing passwords on "important" accounts and that individual meter design decisions likely have a marginal impact.

The Second Usable Privacy and Security for Mobile Devices Workshop (U-PriSM 2) was held with MobileHCI'13. The U-PriSM 2 workshop was an opportunity for researchers and practitioners to discuss research challenges and experiences around the usable privacy and security of mobile devices (smart phones and tablets). Security often involves having non-security experts, or even novice users, regularly making important security decisions while their main focus is on other primary tasks. This is especially true for mobile devices where users can quickly and easily install apps, where user interfaces are minimal due to space constraints, and where users are often distracted by their environment.

A New Security Paradigms Workshop (2013) panel discussed the topic of ethical issues and implications related to markets for zero-day exploits, i.e., markets facilitating the sale of previously unknown details on how to exploit software vulnerabilities in target applications or systems. The related topic of vulnerability rewards programs ("bug bounties" offered by software vendors) was also discussed. This note provides selected background material submitted prior to the panel presentation, and summarizes discussion resulting from the input of both the panelists and NSPW participants.

Smartphone applications pose interesting security problems because the same resources they use to enhance the user experience may also be used in ways that users might find objectionable. We performed a set of experiments to study whether attribution mechanisms could help users understand how smartphone applications access device resources. First, we performed an online survey and found that, as attribution mechanisms have become available on the Android platform, users notice and use them. Second, we designed new attribution mechanisms; a qualitative experiment suggested that our proposed mechanisms are intuitive to understand. Finally, we performed a laboratory experiment in which we simulated application misbehaviors to observe whether users equipped with our attribution mechanisms were able to identify the offending applications. Our results show that, for users who notice application misbehaviors, these attribution mechanisms are significantly more effective than the status quo.

Android's permission system is intended to inform users about the risks of installing applications. When a user installs an application, he or she has the opportunity to review the application's permission requests and cancel the installation if the permissions are excessive or objectionable. We examine whether the Android permission system is effective at warning users. In particular, we evaluate whether Android users pay attention to, understand, and act on permission information during installation. We performed two usability studies: an Internet survey of 308 Android users, and a laboratory study wherein we interviewed and observed 25 Android users. Study participants displayed low attention and comprehension rates: both the Internet survey and laboratory study found that 17% of participants paid attention to permissions during installation, and only 3% of Internet survey respondents could correctly answer all three permission comprehension questions. This indicates that current Android permission warnings do not help most users make correct security decisions. However, a notable minority of users demonstrated both awareness of permission warnings and reasonable rates of comprehension. We present recommendations for improving user attention and comprehension, as well as identify open challenges.

We measure users' attitudes toward interpersonal privacy concerns on Facebook and measure users' strategies for reconciling their concerns with their desire to share content online. To do this, we recruited 260 Facebook users to install a Facebook application that surveyed their privacy concerns, their friend network compositions, the sensitivity of posted content, and their privacy-preserving strategies. By asking participants targeted questions about people randomly selected from their friend network and posts shared on their profiles, we were able to quantify the extent to which users trust their "friends" and the likelihood that their content was being viewed by unintended audiences. We found that while strangers are the most concerning audience, almost 95% of our participants had taken steps to mitigate those concerns. At the same time, we observed that 16.5% of participants had at least one post that they were uncomfortable sharing with a specific friend---someone who likely already had the ability to view it---and that 37% raised more general concerns with sharing their content with friends. We conclude that the current privacy controls allow users to effectively manage the outsider threat, but that they are unsuitable for mitigating concerns over the insider threat---members of the friend network who dynamically become inappropriate audiences based on the context of a post.

We performed a study of Facebook users to examine how they coped with limitations of the Facebook privacy settings interface. Students graduating and joining the workforce create significant problems for all but the most basic privacy settings on social networking websites. We therefore created realistic scenarios exploiting work/play boundaries that required users to specify access control policies that were impossible due to various limitations. We examined whether users were aware of these problems without being prompted, and once given feedback, what their coping strategies were. Overall, we found that simply alerting participants to potential errors was ineffective, but when choices were also presented, participants introduced significantly fewer errors. Based on our findings, we designed a privacy settings interface based on Venn diagrams, which we validated with a usability study. We conclude that this interface may be more effective than the current privacy settings interface.

Text-based passwords are the most common mechanism for authenticating humans to computer systems. To prevent users from picking passwords that are too easy for an adversary to guess, system administrators adopt password-composition policies (e.g., requiring passwords to contain symbols and numbers). Unfortunately, little is known about the relationship between password-composition policies and the strength of the resulting passwords, or about the behavior of users (e.g., writing down passwords) in response to different policies. We present a large-scale study that investigates password strength, user behavior, and user sentiment across four password-composition policies. We characterize the predictability of passwords by calculating their entropy, and find that a number of commonly held beliefs about password composition and strength are inaccurate. We correlate our results with user behavior and sentiment to produce several recommendations for password-composition policies that result in strong passwords without unduly burdening users.

Researchers have recently begun to study the economics of the markets for illicit digital goods to better understand how to invest resources in the most effective mitigations. This line of work in security economics can greatly benefit from data gathering methodologies used for the study of another underground economy, which has been analyzed for the better part of a century: the illicit drug trade. We describe promises and puzzles in the use of observational data for computer security research, that have been encountered previously in drug policy research, and highlight possible lessons we can learn from this different field. We then outline potential opportunities for security research to avoid pitfalls in data collection that drug policy studies have uncovered. Finally, we argue that failure to tackle problems with observational data runs the risk of creating incorrect mythical numbers that can have lasting effects on public policy surrounding computer security.

Many commerce websites post privacy policies to address Internet shoppers' privacy concerns. However, few users read or understand them. Iconic privacy indicators may make privacy policies more accessible and easier for users to understand: in this paper, we examine whether the timing and placement of online privacy indicators impact Internet users' browsing and purchasing decisions. We conducted a laboratory study where we controlled the placement of privacy information, the timing of its appearance, the privacy level of each website, and the price and items being purchased. We found that the timing of privacy information had a significant impact on how much of a premium users were willing to pay for privacy. We also found that timing had less impact when users were willing to examine multiple websites. Finally, we found that users paid more attention to privacy indicators when purchasing privacy-sensitive items than when purchasing items that raised minimal privacy concerns.

Backup authentication mechanisms help users who have forgotten their passwords regain access to their accounts-or at least try. Today's systems fall short in meeting both security and reliability requirements. We designed, built, and tested a new backup authentication system that employs a social-authentication mechanism. The system employs trustees previously appointed by the account holder to verify the account holder's identity. We ran three experiments to determine whether the system could (1) reliably authenticate account holders, (2) resist email attacks that target trustees by impersonating account holders, and (3) resist phone-based attacks from individuals close to account holders. Results were encouraging: seventeen of the nineteen participants who made the effort to call trustees authenticated successfully. However, we also found that users must be reminded of who their trustees are. While email-based attacks were largely unsuccessful, stronger countermeasures will be required to counter highly-personalized phone-based attacks.

All four of the most popular webmail providers ‐ AOL, Google, Microsoft, and Yahoo! ‐ rely on personal questions as the secondary authentication secrets used to reset account passwords. The security of these questions has received limited formal scrutiny, almost all of which predates webmail. We ran a user study to measure the reliability and security of the questions used by all four webmail providers. We asked participants to answer these questions and then asked their acquaintances to guess their answers. Acquaintances with whom participants reported being unwilling to share their webmail passwords were able to guess 17% of their answers. Participants forgot 20% of their own answers within six months. What’s more, 13% of answers could be guessed within five attempts by guessing the most popular answers of other participants, though this weakness is partially attributable to the geographic homogeneity of our participant pool.

Backup authentication mechanisms help users who have forgotten their passwords regain access to their accounts-or at least try. Today's systems fall short in meeting both security and reliability requirements. We designed, built, and tested a new backup authentication system that employs a social-authentication mechanism. The system employs trustees previously appointed by the account holder to verify the account holder's identity. We ran three experiments to determine whether the system could (1) reliably authenticate account holders, (2) resist email attacks that target trustees by impersonating account holders, and (3) resist phone-based attacks from individuals close to account holders. Results were encouraging: seventeen of the nineteen participants who made the effort to call trustees authenticated successfully. However, we also found that users must be reminded of who their trustees are. While email-based attacks were largely unsuccessful, stronger countermeasures will be required to counter highly-personalized phone-based attacks.

No abstract available.

All four of the most popular webmail providers - AOL, Google, Microsoft, and Yahoo! - rely on personal questions as the secondary authentication secrets used to reset account passwords. The security of these questions has received limited formal scrutiny, almost all of which predates webmail. We ran a user study to measure the reliability and security of the questions used by all four webmail providers. We asked participants to answer these questions and then asked their acquaintances to guess their answers. Acquaintances with whom participants reported being unwilling to share their webmail passwords were able to guess 17% of their answers. Participants forgot 20% of their own answers within six months. What's more, 13% of answers could be guessed within five attempts by guessing the most popular answers of other participants, though this weakness is partially attributable to the geographic homogeneity of our participant pool.

Many popular web browsers are now including active phishing warnings after previous research has shown that passive warnings are often ignored. In this laboratory study we examine the effectiveness of these warnings and examine if, how, and why they fail users. We simulated a spear phishing attack to expose users to browser warnings. We found that 97% of our sixty participants fell for at least one of the phishing messages that we sent them. However, we also found that when presented with the active warnings, 79% of participants heeded them, which was not the case for the passive warning that we tested---where only one participant heeded the warnings. Using a model from the warning sciences we analyzed how users perceive warning messages and offer suggestions for creating more effective warning messages within the phishing context.

In this paper we present Family Accounts, a new user account model for shared home computers. We conducted a study with sixteen families, eight who used individual profiles at home, and eight who shared a single profile. Our results demonstrate that Family Accounts is a good compromise between a single shared profile and individual profiles for each family member. In particular, we observed that because Family Accounts allowed individuals to switch profiles without forcing them to interrupt their tasks, family members tended to switch to their own profiles only when a task required some degree of privacy or personalization.

Search engines play an important role in helping users find desired content. With the increasing deployment of computerreadable privacy policies encoded using the standard W3C Platform for Privacy Preferences (P3P) format, search engines also have the potential to help users identify web sites that will respect their privacy needs. We conducted a study of the quantity and quality of P3P-encoded privacy policies associated with top-20 search results from three popular search engines. We built a P3P-enabled search engine and used it to gather statistics on P3P adoption as well as the privacy landscape of the Internet as a whole. This search engine makes use of a privacy policy cache that we designed to facilitate fast searches. Using a list of “typical ” search terms taken from AOL users ’ queries, we examined the trends in privacy policies that are returned from queries to the AOL, Google, and Yahoo! search engines. We then compared these results to results compiled after using “e-commerce search terms ” from Google’s Froogle service. We examined the top 20 search results returned by each search engine for each of the search terms and found at least one result with a P3P policy for 83 % of the typical search terms. Overall we found that these typical search terms yielded P3P adoption rates of 10%. This contrasts with adoption rates of 21 % percent when searching for e-commerce terms. Examining the content of the policies, we discovered that a minority of sites engage in direct marketing with or without a way of opting out, and that even fewer sites share personal information with other companies. Finally, we outline ways to increase P3P adoption rates as well as decrease policy errors.

While Internet users claim to be concerned about online privacy, their behavior rarely reflects those concerns. In this paper we investigate whether the availability of comparison information about the privacy practices of online merchants affects users’ behavior. We conducted our study using Privacy Finder, a “privacy-enhanced search engine” that displays search results annotated with the privacy policy information of each site. The privacy information is garnered from computer-readable privacy policies found at the respective sites. We asked users to purchase one nonprivacy- sensitive item and then one privacy-sensitive item using Privacy Finder, and observed whether the privacy information provided by our search engine impacted users’ purchasing decisions (participants’ costs were reimbursed, in order to separate the effect of privacy policies from that of price). A control group was asked to make the same purchases using a search engine that produced the same results as Privacy Finder, but did not display privacy information. We found that while Privacy Finder had some influence on non-privacy-sensitive purchase decisions, it had a more significant impact on privacy-sensitive purchases. The results suggest that when privacy policy comparison information is readily available, individuals may be willing to seek out more privacy friendly web sites and perhaps even pay a premium for privacy depending on the nature of the items to be purchased.