The surge of spreading harmful information through digital technologies: a distressing reality in complex humanitarian emergencies

Digital technologies, such as social media and artificial intelligence (AI), are increasingly used to spread harmful information during complex humanitarian emergencies. Spreading false or harmful information during emergencies is not new, but the increase in the quantity of harmful information in the last decade is unprecedented.1Adams Z Osman M Bechlivanidis C Meder B (Why) is misinformation a problem?.Perspect Psychol Sci. 2023; (published online Feb 16.)https://doi.org/10.1177/17456916221141344Crossref PubMed Scopus (0) Google Scholar Numerous allegations and accusations about the International Committee of the Red Cross and other humanitarian organisations, disseminated widely through different online platforms during recent conflicts, highlight this surge in spreading harmful information. The video known as cash-in-trunks was shared widely in conflict settings such as Syria and Ukraine, showing boxes with the organisation's logo loaded with money, accusing the International Committee of the Red Cross of illegal and corrupt practices.2International Committee of the Red CrossFalse allegations: ICRC condemns video showing cash in trunks. International Committee of the Red Cross.https://www.icrc.org/en/document/false-allegations-icrc-condemns-video-cash-trunksDate: 2019Date accessed: March 23, 2023Google Scholar The 2018 accusations, claiming that Save the Children colluded with human traffickers while rescuing displaced people in the Mediterranean, is another example.3Bunce M Humanitarian communication in a post-truth world.Journal of Humanitarian Affairs. 2019; 1: 49-55Crossref Google Scholar These false narratives give the impression that humanitarian organisations have violated international humanitarian law, international human rights law, and international criminal law. This sort of false information threatens the ability of agencies to address complex humanitarian emergencies safely and effectively. Spreading harmful information, including but not limited to disinformation (false information disseminated with malicious intent), malinformation (true information but intentionally spread to cause harm), and misinformation (false information spread by individuals who believe the information to be true), could exacerbate the suffering of the most vulnerable people, and humanitarian operations. Harmful information is reported to induce psychological and social harm in both communities affected by conflict and among people serving those communities.4Tran T, Valecha R, Rad P, Rao RH. Taxonomy of misinformation harms from social media in humanitarian crises. American Conference in Information Systems 2019; Aug 15–17, 2019 (abstr 92).Google Scholar Incorrect information prevents communities experiencing humanitarian emergencies from accessing important and lifesaving services, and undermines the ability of humanitarian organisations to deliver and implement effective interventions. A WHO review reported that spreading false information increases delays in providing needed interventions during emergencies, and damages people's mental health.5Borges do Nascimento IJ Pizarro AB Almeida JM et al.Infodemics and health misinformation: a systematic review of reviews.Bull World Health Organ. 2022; 100: 544-561Crossref PubMed Scopus (8) Google Scholar However, the long-term effects of these harmful practices have not been studied in either people who have experienced humanitarian emergencies, or in humanitarian aid workers. Complex humanitarian settings have become fertile environments for spreading misinformation, disinformation, and malinformation. Weak governance systems have limited ability to confront and prevent the dissemination of digitally generated materials.6Katz A Liar's war: protecting civilians from disinformation during armed conflict.Int Rev Red Cross. 2020; 102: 659-682Crossref Scopus (0) Google Scholar It is especially difficult for those dealing with humanitarian emergencies to distinguish between false or harmful information and beneficial messages, given the plethora of information available on social media platforms. For example, in Iraq, social media platforms were accessed by 28 million users in 2022, ie, 67% of the 42 million citizens, compared with 25 million users in the previous year.7Iraqi News Agency28 million users of social networking sites in Iraq.https://ina.iq/eng/17634-dmc-28-million-users-of-social-networking-sites-in-iraq.htmlDate: Feb 19, 2022Date accessed: March 23, 2023Google Scholar This growth in usage contributes to the increased exposure and consumption of incorrect information. Studies reported that the spread of video-based misinformation or disinformation (which is more likely to mislead and go viral compared with text and image formats, as it is difficult to detect) is rapidly increasing. These videos could be easily weaponised to incite conflict and unrest, just as they were used successfully by ISIS during its radicalisation and recruitment campaigns in Iraq and Syria.8Bu Y Sheng Q Cao J Qi P Wang D Li J Online misinformation video detection: a survey.arXiv. 2023; (published online Feb 7.) (preprint).https://doi.org/10.48550/arXiv.2302.03242Google Scholar Timely detection of these dangerous materials relies on labour-intensive and time-consuming procedures that are prone to bias and errors, which contribute to failures in performing real-time detection. Improper use of AI also poses a high risk in complex humanitarian emergencies. For example, a 2021 study reported that AI systems can generate and spread automated misinformation, convincing enough to fool and mislead cybersecurity and other experts, without further human involvement after an initial prompt.9Ranade P Piplai A Mittal S Joshi A Finin T Generating fake cyber threat intelligence using transformer-based models.arXiv. 2021; (published online Feb 8.) (preprint).https://doi.org/10.48550/arXiv.2102.04351Google Scholar Social media platforms that use AI-powered algorithms can generate inflammatory content that is consumed by users, and possibly contribute to an increase in violence and conflict. At present, ethical guidance or frameworks on AI ethics in humanitarian settings are scarce. Collaboration is required to stop the spread of misleading or falsified information. The UN #PledgetoPause campaign, implemented during the COVID-19 pandemic, found that taking a brief pause before sharing information can substantially reduce the tendency to share materials designed to elicit a strong emotional reaction, and thus slow down their spread.10UNUN urges people to #pledgetopause before sharing information.https://news.un.org/en/story/2020/10/1075742Date: Oct 21, 2020Date accessed: March 23, 2023Google Scholar This approach could be replicated during complex humanitarian emergencies. A global strategy is needed to address this issue. The 2021 release of the Oxford Statement on International Law Protection in Cyberspace, which touches on sovereignty, incitement, human rights, criminal law, general rules of international humanitarian law, and international criminal law, is an important first step to address disinformation.6Katz A Liar's war: protecting civilians from disinformation during armed conflict.Int Rev Red Cross. 2020; 102: 659-682Crossref Scopus (0) Google Scholar Those implementing these programmes should identify the sources and hold the creators accountable for the damage inflicted. In addition, online platforms such as social media and news websites, should implement effective measures to prevent the spread of potentially harmful information through their outlets or platforms. The effects of spreading harmful information should be thoroughly considered when responding to complex humanitarian emergencies. Understanding the dimensions of this multifaceted problem can enable humanitarian aid agencies to address the challenges and apply best practices to facilitate an effective humanitarian response. I declare no competing interests.