We applied a mixed-methods approach with the goal of understanding how Latinx and Asian diaspora communities perceive and experience the spread of misinformation through encrypted messaging apps in the United States. Our study consists of 12 in-depth interviews with leaders of relevant diaspora community organizations and a computer-assisted content analysis of 450,300 messages published on Telegram between July 2020 and December 2021. We found evidence of cross-platform misinformation sharing, particularly between Telegram, WhatsApp, and YouTube. The enclosed nature of encrypted messaging applications makes them a testing ground for misinformation narratives before these narratives are sent out to open platforms. Finally, YouTube is a central component of misinformation spread because much of the misinformation content spread in these communities is video-based.
{"title":"Taking the power back: How diaspora community organizations are fighting misinformation spread on encrypted messaging apps","authors":"J. V. S. Ozawa, Samuel Woolley, Josephine Lukito","doi":"10.37016/mr-2020-146","DOIUrl":"https://doi.org/10.37016/mr-2020-146","url":null,"abstract":"We applied a mixed-methods approach with the goal of understanding how Latinx and Asian diaspora communities perceive and experience the spread of misinformation through encrypted messaging apps in the United States. Our study consists of 12 in-depth interviews with leaders of relevant diaspora community organizations and a computer-assisted content analysis of 450,300 messages published on Telegram between July 2020 and December 2021. We found evidence of cross-platform misinformation sharing, particularly between Telegram, WhatsApp, and YouTube. The enclosed nature of encrypted messaging applications makes them a testing ground for misinformation narratives before these narratives are sent out to open platforms. Finally, YouTube is a central component of misinformation spread because much of the misinformation content spread in these communities is video-based.","PeriodicalId":93289,"journal":{"name":"Harvard Kennedy School misinformation review","volume":"24 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141353955","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Rongwei Tang, E. Vraga, Leticia Bode, Shelley Boulianne
Observed corrections of misinformation on social media can encourage more accurate beliefs, but for these benefits to occur, corrections must happen. By exploring people’s perceptions of witnessing and performing corrections on social media, we find that many people say they observe and perform corrections across the United States, the United Kingdom, Canada, and France. We find higher levels of self-reported correction experiences in the United States but few differences between who reports these experiences across countries. Specifically, younger and more educated adults, as well as those who see misinformation more frequently online, are more likely to report observing and performing corrections across contexts.
{"title":"Who reports witnessing and performing corrections on social media in the United States, United Kingdom, Canada, and France?","authors":"Rongwei Tang, E. Vraga, Leticia Bode, Shelley Boulianne","doi":"10.37016/mr-2020-145","DOIUrl":"https://doi.org/10.37016/mr-2020-145","url":null,"abstract":"Observed corrections of misinformation on social media can encourage more accurate beliefs, but for these benefits to occur, corrections must happen. By exploring people’s perceptions of witnessing and performing corrections on social media, we find that many people say they observe and perform corrections across the United States, the United Kingdom, Canada, and France. We find higher levels of self-reported correction experiences in the United States but few differences between who reports these experiences across countries. Specifically, younger and more educated adults, as well as those who see misinformation more frequently online, are more likely to report observing and performing corrections across contexts.","PeriodicalId":93289,"journal":{"name":"Harvard Kennedy School misinformation review","volume":"1 7","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141265723","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Generative artificial intelligence (AI) models have introduced new complexities and risks to information environments, as synthetic media may facilitate the spread of misinformation and erode public trust. This study examines the prevalence and characteristics of synthetic media on social media platform X from December 2022 to September 2023. Leveraging crowdsourced annotations identifying synthetic content, our analysis reveals an increase in AI-generated media over time, with an initial spike in March 2023, following the release of Midjourney V5. While most synthetic media identified is non-political and non-malicious, concerning deepfakes targeting political figures persist, raising questions on the potential for misuse of AI technologies.
{"title":"The spread of synthetic media on X","authors":"Giulio Corsi, Bill Marino, Willow Wong","doi":"10.37016/mr-2020-140","DOIUrl":"https://doi.org/10.37016/mr-2020-140","url":null,"abstract":"Generative artificial intelligence (AI) models have introduced new complexities and risks to information environments, as synthetic media may facilitate the spread of misinformation and erode public trust. This study examines the prevalence and characteristics of synthetic media on social media platform X from December 2022 to September 2023. Leveraging crowdsourced annotations identifying synthetic content, our analysis reveals an increase in AI-generated media over time, with an initial spike in March 2023, following the release of Midjourney V5. While most synthetic media identified is non-political and non-malicious, concerning deepfakes targeting political figures persist, raising questions on the potential for misuse of AI technologies.","PeriodicalId":93289,"journal":{"name":"Harvard Kennedy School misinformation review","volume":"50 21","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141269714","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In a preliminary analysis of 121,984 posts from X (formerly known as Twitter) containing the hashtag #SaveTheChildren, we found that conspiratorial posts received more engagement than authentic hashtag activism between January 2022 and March 2023. Conspiratorial posts received twice the number of reposts as non-conspiratorial content. Our initial findings of a forthcoming larger multi-platform study suggest that the way that users strategically exploit the #SaveTheChildren hashtag may have implications for the visibility of legitimate social movements on the X platform. Future work should consider other social media platforms to determine if the visibility of legitimate social movements is decreasing more broadly, particularly in the context of 2024 being the largest election year in history.
{"title":"#SaveTheChildren: A pilot study of a social media movement co-opted by conspiracy theorists","authors":"Katherine M. FitzGerald, Timothy Graham","doi":"10.37016/mr-2020-143","DOIUrl":"https://doi.org/10.37016/mr-2020-143","url":null,"abstract":"In a preliminary analysis of 121,984 posts from X (formerly known as Twitter) containing the hashtag #SaveTheChildren, we found that conspiratorial posts received more engagement than authentic hashtag activism between January 2022 and March 2023. Conspiratorial posts received twice the number of reposts as non-conspiratorial content. Our initial findings of a forthcoming larger multi-platform study suggest that the way that users strategically exploit the #SaveTheChildren hashtag may have implications for the visibility of legitimate social movements on the X platform. Future work should consider other social media platforms to determine if the visibility of legitimate social movements is decreasing more broadly, particularly in the context of 2024 being the largest election year in history.","PeriodicalId":93289,"journal":{"name":"Harvard Kennedy School misinformation review","volume":"6 12","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141104537","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Ho-Chun Herbert Chang, Austin Horng-En Wang, Yu Sunny Fang
Taiwan has one of the highest freedom of speech indexes while it also encounters the largest amount of foreign interference due to its contentious history with China. Because of the large influx of misinformation, Taiwan has taken a public crowdsourcing approach to combatting misinformation, using both fact-checking ChatBots and public dataset called CoFacts. Combining CoFacts with large-language models (LLM), we investigated misinformation across three platforms (Line, PTT, and Facebook) during the 2024 Taiwanese presidential elections. We found that most misinformation appears within China-friendly political groups and attacks US-Taiwan relations through visual media like images and videos. A considerable proportion of misinformation does not question U.S. foreign policy directly. Rather, it exaggerates domestic issues in the United States to create a sense of declining U.S. state capacity. Curiously, we found misinformation rhetoric that references conspiracy groups in the West.
{"title":"US-skepticism and transnational conspiracy in the 2024 Taiwanese presidential election","authors":"Ho-Chun Herbert Chang, Austin Horng-En Wang, Yu Sunny Fang","doi":"10.37016/mr-2020-144","DOIUrl":"https://doi.org/10.37016/mr-2020-144","url":null,"abstract":"Taiwan has one of the highest freedom of speech indexes while it also encounters the largest amount of foreign interference due to its contentious history with China. Because of the large influx of misinformation, Taiwan has taken a public crowdsourcing approach to combatting misinformation, using both fact-checking ChatBots and public dataset called CoFacts. Combining CoFacts with large-language models (LLM), we investigated misinformation across three platforms (Line, PTT, and Facebook) during the 2024 Taiwanese presidential elections. We found that most misinformation appears within China-friendly political groups and attacks US-Taiwan relations through visual media like images and videos. A considerable proportion of misinformation does not question U.S. foreign policy directly. Rather, it exaggerates domestic issues in the United States to create a sense of declining U.S. state capacity. Curiously, we found misinformation rhetoric that references conspiracy groups in the West.","PeriodicalId":93289,"journal":{"name":"Harvard Kennedy School misinformation review","volume":"6 21","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141120289","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This study integrates research on negativity bias and misinformation, as a comparison of how systematic (negativity) and incidental (misinformation) challenges to the news are perceived differently by audiences. Through a cross-country survey, we found that both challenges are perceived as highly salient and disruptive. Despite negativity bias in the news possibly being a more widespread phenomenon, respondents across the surveyed countries perceive misinformation as a relatively bigger threat, even in countries where negativity is estimated to be more prevalent. In conclusion, the optimism of recent research about people's limited misinformation exposure does not seem to be reflected in audiences’ threat perceptions.
{"title":"Misinformation perceived as a bigger informational threat than negativity: A cross-country survey on challenges of the news environment","authors":"T. G. van der Meer, M. Hameleers","doi":"10.37016/mr-2020-142","DOIUrl":"https://doi.org/10.37016/mr-2020-142","url":null,"abstract":"This study integrates research on negativity bias and misinformation, as a comparison of how systematic (negativity) and incidental (misinformation) challenges to the news are perceived differently by audiences. Through a cross-country survey, we found that both challenges are perceived as highly salient and disruptive. Despite negativity bias in the news possibly being a more widespread phenomenon, respondents across the surveyed countries perceive misinformation as a relatively bigger threat, even in countries where negativity is estimated to be more prevalent. In conclusion, the optimism of recent research about people's limited misinformation exposure does not seem to be reflected in audiences’ threat perceptions.","PeriodicalId":93289,"journal":{"name":"Harvard Kennedy School misinformation review","volume":"15 6","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140981757","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
While algorithms and crowdsourcing have been increasingly used to debunk or label misinformation on social media, such tasks might be most effective when performed by professional fact checkers or journalists. Drawing on a national survey (N = 1,003), we found that U.S. adults evaluated fact-checking labels created by professional fact-checkers as more effective than labels by algorithms and other users. News media labels were perceived as more effective than user labels but not statistically different from labels by fact checkers and algorithms. There was no significant difference between labels created by users and algorithms. These findings have implications for platforms and fact-checking practitioners, underscoring the importance of journalistic professionalism in fact-checking.
{"title":"Journalistic interventions matter: Understanding how Americans perceive fact-checking labels","authors":"Chenyan Jia, Taeyoung Lee","doi":"10.37016/mr-2020-138","DOIUrl":"https://doi.org/10.37016/mr-2020-138","url":null,"abstract":"While algorithms and crowdsourcing have been increasingly used to debunk or label misinformation on social media, such tasks might be most effective when performed by professional fact checkers or journalists. Drawing on a national survey (N = 1,003), we found that U.S. adults evaluated fact-checking labels created by professional fact-checkers as more effective than labels by algorithms and other users. News media labels were perceived as more effective than user labels but not statistically different from labels by fact checkers and algorithms. There was no significant difference between labels created by users and algorithms. These findings have implications for platforms and fact-checking practitioners, underscoring the importance of journalistic professionalism in fact-checking.","PeriodicalId":93289,"journal":{"name":"Harvard Kennedy School misinformation review","volume":"49 6","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140713610","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
J. V. S. Ozawa, Josephine Lukito, Felipe Bailez, Luis G. P. Fakhouri
Bolsonaro’s supporters used social media to spread content during key events related to the Brasília attack. An unprecedented analysis of more than 15,000 public WhatsApp groups showed that these political actors tried to manufacture consensus in preparation for and after the attack. A cross-platform time series analysis showed that the spread of content on Twitter predicted the spread of content on WhatsApp. Twitter also predicted news coverage of Bolsonaro’s supporters, suggesting a propaganda feedback loop. Our findings indicate that investigative journalism and public policy initiatives could benefit from monitoring public groups on encrypted messaging apps.
{"title":"Brazilian Capitol attack: The interaction between Bolsonaro’s supporters’ content, WhatsApp, Twitter, and news media","authors":"J. V. S. Ozawa, Josephine Lukito, Felipe Bailez, Luis G. P. Fakhouri","doi":"10.37016/mr-2020-137","DOIUrl":"https://doi.org/10.37016/mr-2020-137","url":null,"abstract":"Bolsonaro’s supporters used social media to spread content during key events related to the Brasília attack. An unprecedented analysis of more than 15,000 public WhatsApp groups showed that these political actors tried to manufacture consensus in preparation for and after the attack. A cross-platform time series analysis showed that the spread of content on Twitter predicted the spread of content on WhatsApp. Twitter also predicted news coverage of Bolsonaro’s supporters, suggesting a propaganda feedback loop. Our findings indicate that investigative journalism and public policy initiatives could benefit from monitoring public groups on encrypted messaging apps.","PeriodicalId":93289,"journal":{"name":"Harvard Kennedy School misinformation review","volume":"57 S13","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140721584","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Statements of fact can be proved or disproved with objective evidence, whereas statements of opinion depend on personal values and preferences. Distinguishing between these types of statements contributes to information competence. Conversely, failure at fact-opinion differentiation potentially brings resistance to corrections of misinformation and susceptibility to manipulation. Our analyses show that on fact-opinion differentiation tasks, unsystematic mistakes and mistakes emanating from partisan bias occur at higher rates than accurate responses. Accuracy increases with political sophistication. Affective partisan polarization promotes systematic partisan error: As views grow more polarized, partisans increasingly see their side as holding facts and the opposing side as holding opinions.
{"title":"Fact-opinion differentiation","authors":"Matthew Mettler, Jeffery J. Mondak","doi":"10.37016/mr-2020-136","DOIUrl":"https://doi.org/10.37016/mr-2020-136","url":null,"abstract":"Statements of fact can be proved or disproved with objective evidence, whereas statements of opinion depend on personal values and preferences. Distinguishing between these types of statements contributes to information competence. Conversely, failure at fact-opinion differentiation potentially brings resistance to corrections of misinformation and susceptibility to manipulation. Our analyses show that on fact-opinion differentiation tasks, unsystematic mistakes and mistakes emanating from partisan bias occur at higher rates than accurate responses. Accuracy increases with political sophistication. Affective partisan polarization promotes systematic partisan error: As views grow more polarized, partisans increasingly see their side as holding facts and the opposing side as holding opinions.","PeriodicalId":93289,"journal":{"name":"Harvard Kennedy School misinformation review","volume":"24 12","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140259798","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Johannes Christiern Santos Okholm, Amir Ebrahimi Fard, Marijn ten Thij
Through an online field experiment, we test traditional and novel counter-misinformation strategies among fringe communities. Though generally effective, traditional strategies have not been tested in fringe communities, and do not address the online infrastructure of misinformation sources supporting such consumption. Instead, we propose to activate source criticism by exposing sources’ unreliability. Based on a snowball sampling of German fringe communities on Facebook, we test if debunking and source exposure reduce groups’ consumption levels of two popular misinformation sources. Results support a proactively engaging counter-misinformation approach to reduce consumption of misinformation sources.
{"title":"Debunking and exposing misinformation among fringe communities: Testing source exposure and debunking anti-Ukrainian misinformation among German fringe communities","authors":"Johannes Christiern Santos Okholm, Amir Ebrahimi Fard, Marijn ten Thij","doi":"10.37016/mr-2020-134","DOIUrl":"https://doi.org/10.37016/mr-2020-134","url":null,"abstract":"Through an online field experiment, we test traditional and novel counter-misinformation strategies among fringe communities. Though generally effective, traditional strategies have not been tested in fringe communities, and do not address the online infrastructure of misinformation sources supporting such consumption. Instead, we propose to activate source criticism by exposing sources’ unreliability. Based on a snowball sampling of German fringe communities on Facebook, we test if debunking and source exposure reduce groups’ consumption levels of two popular misinformation sources. Results support a proactively engaging counter-misinformation approach to reduce consumption of misinformation sources.","PeriodicalId":93289,"journal":{"name":"Harvard Kennedy School misinformation review","volume":"9 7","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139958646","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：