arXiv subjects

Emma Pieroni

Publications and source records attributed to Emma Pieroni.

12 recordsLinked to original sources

arXiv2023

Talking Abortion (Mis)information with ChatGPT on TikTok

In this study, we tested users' perception of accuracy and engagement with TikTok videos in which ChatGPT responded to prompts about "at-home" abortion remedies. The chatbot's responses, though somewhat vague and confusing, nonetheless recommended consulting with health professionals before attempting an "at-home" abortion. We used ChatGPT to create two TikTok video variants - one where users can see ChatGPT explicitly typing back a response, and one where the text response is presented without any notion to the chatbot. We randomly exposed 100 participants to each variant and found that the group of participants unaware of ChatGPT's text synthetization was more inclined to believe the responses were misinformation. Under the same impression, TikTok itself attached misinformation warning labels ("Get the facts about abortion") to all videos after we collected our initial results. We then decided to test the videos again with another set of 50 participants and found that the labels did not affect the perceptions of abortion misinformation except in the case where ChatGPT explicitly responded to a prompt for a lyrical output. We also found that more than 60% of the participants expressed negative or hesitant opinions about chatbots as sources of credible health information.

cs.HC↗

arXiv2023

Abortion Misinformation on TikTok: Rampant Content, Lax Moderation, and Vivid User Experiences

The scientific effort devoted to health misinformation mostly focuses on the implications of misleading vaccines and communicable disease claims with respect to public health. However, the proliferation of abortion misinformation following the Supreme Court's decision to overturn Roe v. Wade banning legal abortion in the US highlighted a gap in scientific attention to individual health-related misinformation. To address this gap, we conducted a study with 60 TikTok users to uncover their experiences with abortion misinformation and the way they conceptualize, assess, and respond to misleading video content on this platform. Our findings indicate that users mostly encounter short-term videos suggesting herbal "at-home" remedies for pregnancy termination. While many of the participants were cautious about scientifically debunked "abortion alternatives," roughly 30% of the entire sample believed in their safety and efficacy. Even an explicit debunking label attached to a misleading abortion video about the harms of "at-home" did not help a third of the participants to dismiss a video about self-administering abortion as misinformation. We discuss the implications of our findings for future participation on TikTok and other polarizing topics debated on social media.

cs.SI↗

arXiv2022

Folk Models of Misinformation on Social Media

In this paper we investigate what folk models of misinformation exist through semi-structured interviews with a sample of 235 social media users. Work on social media misinformation does not investigate how ordinary users - the target of misinformation - deal with it; rather, the focus is mostly on the anxiety, tensions, or divisions misinformation creates. Studying the aspects of creation, diffusion and amplification also overlooks how misinformation is internalized by users on social media and thus is quick to prescribe "inoculation" strategies for the presumed lack of immunity to misinformation. How users grapple with social media content to develop "natural immunity" as a precursor to misinformation resilience remains an open question. We have identified at least five folk models that conceptualize misinformation as either: political (counter)argumentation, out-of-context narratives, inherently fallacious information, external propaganda, or simply entertainment. We use the rich conceptualizations embodied in these folk models to uncover how social media users minimize adverse reactions to misinformation encounters in their everyday lives.

cs.SI↗

arXiv2022

Meaningful Context, a Red Flag, or Both? Users' Preferences for Enhanced Misinformation Warnings on Twitter

Warning users about misinformation on social media is not a simple usability task. Soft moderation has to balance between debunking falsehoods and avoiding moderation bias while preserving the social media consumption flow. Platforms thus employ minimally distinguishable warning tags with generic text under a suspected misinformation content. This approach resulted in an unfavorable outcome where the warnings "backfired" and users believed the misinformation more, not less. In response, we developed enhancements to the misinformation warnings where users are advised on the context of the information hazard and exposed to standard warning iconography. We ran an A/B evaluation with the Twitter's original warning tags in a 337 participant usability study. The majority of the participants preferred the enhancements as a nudge toward recognizing and avoiding misinformation. The enhanced warning tags were most favored by the politically left-leaning and to a lesser degree moderate participants, but they also appealed to roughly a third of the right-leaning participants. The education level was the only demographic factor shaping participants' preferences. We use our findings to propose user-tailored improvements in the soft moderation of misinformation on social media.

cs.CY↗

arXiv2022

"Gettr-ing" Deep Insights from the Social Network Gettr

As yet another alternative social network, Gettr positions itself as the "marketplace of ideas" where users should expect the truth to emerge without any administrative censorship. We looked deep inside the platform by analyzing it's structure, a sample of 6.8 million posts, and the responses from a sample of 124 Gettr users we interviewed to see if this actually is the case. Administratively, Gettr makes a deliberate attempt to stifle any external evaluation of the platform as collecting data is marred with unpredictable and abrupt changes in their API. Content-wise, Gettr notably hosts pro-Trump content mixed with conspiracy theories and attacks on the perceived "left." It's social network structure is asymmetric and centered around prominent right-thought leaders, which is characteristic for all alt-platforms. While right-leaning users joined Gettr as a result of a perceived freedom of speech infringement by the mainstream platforms, left-leaning users followed them in numbers as to "keep up with the misinformation." We contextualize these findings by looking into the Gettr's user interface design to provide a comprehensive insight into the incentive structure for joining and competing for the truth on Gettr.

cs.SI↗

arXiv2022

Gone Quishing: A Field Study of Phishing with Malicious QR Codes

The COVID-19 pandemic enabled "quishing", or phishing with malicious QR codes, as they became a convenient go-between for sharing URLs, including malicious ones. To explore the quishing phenomenon, we conducted a 173-participant study where we used a COVID-19 digital passport sign-up trial with a malicious QR code as a pretext. We found that 67 % of the participants were happy to sign-up with their Google or Facebook credentials, 18.5% to create a new account, and only 14.5% to skip on the sign-up. Convenience was the single most cited factor for the willingness to yield participants' credentials. Reluctance of linking personal accounts with new services was the reason for creating a new account or skipping the registration. We also developed a Quishing Awareness Scale (QAS) and found a significant relationship between participants' QR code behavior and their sign-up choices: the ones choosing to sign-up with Facebook scored the lowest while the one choosing to skip the highest on average. We used our results to propose quishing awareness training guidelines and develop and test usable security indicators for warning users about the threat of quishing.

cs.CR↗

arXiv2021

(Mis)perceptions and Engagement on Twitter: COVID-19 Vaccine Rumors on Efficacy and Mass Immunization Effort

This paper reports the findings of a 606-participant study where we analyzed the perception and engagement effects of COVID-19 vaccine rumours on Twitter pertaining to (a) vaccine efficacy; and (b) mass immunization efforts in the United States. Misperceptions regarding vaccine efficacy were successfully induced through simple content alterations and the addition of popular anti COVID-19 hashtags to otherwise valid Twitter content. Twitter's misinformation contextual tags caused a "backfire effect" for the skeptic, vaccine-hesitant reinforcing their opposition stance. While the majority of the participants staunchly refrain from engaging with the COVID-19 rumours, the skeptic, vaccine-hesitant ones were open to comment, re-tweet, like and share the vaccine efficacy rumors. We discuss the implications of our results in the context of broadening the effort for dispelling rumors about COVID-19 on social media.

cs.SI↗

arXiv2021

Parlermonium: A Data-Driven UX Design Evaluation of the Parler Platform

This paper evaluates Parler, the controversial social media platform, from two seemingly orthogonal perspectives: UX design perspective and data science. UX design researchers explore how users react to the interface/content of their social media feeds; Data science researchers analyze the misinformation flow in these feeds to detect alternative narratives and state-sponsored disinformation campaigns. We took a critical look into the intersection of these approaches to understand how Parler's interface itself is conductive to the flow of misinformation and the perception of "free speech" among its audience. Parler drew widespread attention leading up to and after the 2020 U.S. elections as the "alternative" place for free speech, as a reaction to other mainstream social media platform which actively engaged in labeling misinformation with content warnings. Because platforms like Parler are disruptive to the social media landscape, we believe the evaluation uniquely uncovers the platform's conductivity to the spread of misinformation.

cs.SI↗

arXiv2021

"Hey Alexa, What do You Know About the COVID-19 Vaccine?" -- (Mis)perceptions of Mass Immunization Among Voice Assistant Users

In this paper, we analyzed the perceived accuracy of COVID-19 vaccine information spoken back by Amazon Alexa. Unlike social media, Amazon Alexa doesn't apply soft moderation to unverified content, allowing for use of third-party malicious skills to arbitrarily phrase COVID-19 vaccine information. The results from a 210-participant study suggest that a third-party malicious skill could successful reduce the perceived accuracy among the users of information as to who gets the vaccine first, vaccine testing, and the side effects of the vaccine. We also found that the vaccine-hesitant participants are drawn to pessimistically rephrased Alexa responses focused on the downsides of the mass immunization. We discuss solutions for soft moderation against misperception-inducing or altogether COVID-19 misinformation malicious third-party skills.

cs.CY↗

arXiv2021

Misinformation Warning Labels: Twitter's Soft Moderation Effects on COVID-19 Vaccine Belief Echoes

Twitter, prompted by the rapid spread of alternative narratives, started actively warning users about the spread of COVID-19 misinformation. This form of soft moderation comes in two forms: as a warning cover before the Tweet is displayed to the user and as a warning tag below the Tweet. This study investigates how each of the soft moderation forms affects the perceived accuracy of COVID-19 vaccine misinformation on Twitter. The results suggest that the warning covers work, but not the tags, in reducing the perception of accuracy of COVID-19 vaccine misinformation on Twitter. "Belief echoes" do exist among Twitter users, unfettered by any warning labels, in relationship to the perceived safety and efficacy of the COVID-19 vaccine as well as the vaccination hesitancy for themselves and their children. The implications of these results are discussed in the context of usable security affordances for combating misinformation on social media.

cs.SI↗

arXiv2021

"TL;DR:" Out-of-Context Adversarial Text Summarization and Hashtag Recommendation

This paper presents Out-of-Context Summarizer, a tool that takes arbitrary public news articles out of context by summarizing them to coherently fit either a liberal- or conservative-leaning agenda. The Out-of-Context Summarizer also suggests hashtag keywords to bolster the polarization of the summary, in case one is inclined to take it to Twitter, Parler or other platforms for trolling. Out-of-Context Summarizer achieved 79% precision and 99% recall when summarizing COVID-19 articles, 93% precision and 93% recall when summarizing politically-centered articles, and 87% precision and 88% recall when taking liberally-biased articles out of context. Summarizing valid sources instead of synthesizing fake text, the Out-of-Context Summarizer could fairly pass the "adversarial disclosure" test, but we didn't take this easy route in our paper. Instead, we used the Out-of-Context Summarizer to push the debate of potential misuse of automated text generation beyond the boilerplate text of responsible disclosure of adversarial language models.

cs.CL↗

arXiv2020

TrollHunter2020: Real-Time Detection of Trolling Narratives on Twitter During the 2020 US Elections

This paper presents TrollHunter2020, a real-time detection mechanism we used to hunt for trolling narratives on Twitter during the 2020 U.S. elections. Trolling narratives form on Twitter as alternative explanations of polarizing events like the 2020 U.S. elections with the goal to conduct information operations or provoke emotional response. Detecting trolling narratives thus is an imperative step to preserve constructive discourse on Twitter and remove an influx of misinformation. Using existing techniques, this takes time and a wealth of data, which, in a rapidly changing election cycle with high stakes, might not be available. To overcome this limitation, we developed TrollHunter2020 to hunt for trolls in real-time with several dozens of trending Twitter topics and hashtags corresponding to the candidates' debates, the election night, and the election aftermath. TrollHunter2020 collects trending data and utilizes a correspondence analysis to detect meaningful relationships between the top nouns and verbs used in constructing trolling narratives while they emerge on Twitter. Our results suggest that the TrollHunter2020 indeed captures the emerging trolling narratives in a very early stage of an unfolding polarizing event. We discuss the utility of TrollHunter2020 for early detection of information operations or trolling and the implications of its use in supporting a constrictive discourse on the platform around polarizing topics.

cs.CR↗