Crisis Informatics and Social Media: Rethinking the Digital Frontier of Emergency Response

Crisis informatics is a field that lives at the crossroads of technology, communication, and emergency management. At its heart lies a deceptively simple question: in the midst of a disaster, can the vast, chaotic streams of social media actually be turned into something useful? When a hurricane bears down on a coast, when a gunman opens fire in a crowded venue, when a virus sweeps silently through a population, millions of people reach for their phones. They post, share, retweet, and comment. The information that emerges is messy, contradictory, and often unreliable. Yet embedded within the noise are signals of extraordinary value—reports from the ground, calls for help, offers of shelter, the location of a safe water source, the name of a road that has become impassable. For more than a decade, researchers in crisis informatics have pursued the dream of harvesting these signals. Their ambition was not merely academic. They wanted to give emergency responders better tools, to help overwhelmed government agencies tap into the collective intelligence of ordinary citizens, and to transform social media from a chaotic cacophony into a structured ecosystem of actionable knowledge. This dream seemed attainable. The tools of data science were advancing, platforms were becoming more open to researchers, and a vibrant community of scholars was building ever more sophisticated methods for tracking, validating, and visualizing crisis-related information. Then came COVID-19, and with it, a profound reckoning. The pandemic tested nearly every assumption the field had been built upon, dismantled its methodological foundations, and forced researchers to confront a set of uncomfortable truths about power, surveillance, and the very nature of digital citizenship.

In the years before the pandemic, crisis informatics was suffused with a certain hopefulness—a belief that social media were fundamentally democratic instruments in times of trouble. Researchers had charted again and again how citizens used platforms like Twitter and Facebook to report emergencies, coordinate rescue efforts, share evacuation routes, and knit together informal networks of mutual aid. The 2010 Haiti earthquake became a landmark case study, as crowdsourced mapping initiatives helped guide relief workers through the chaos. Hurricane Sandy in 2012 offered another powerful example, with New Yorkers turning to social media not just to consume official advisories but to autonomously organize their own systems of care, checking in on the elderly, crowdsourcing everything from warm blankets to working gas stations, and amplifying warnings that sometimes outperformed official alerts. This pattern repeated across disasters in Japan, Australia, and elsewhere, cementing a narrative in the research community of a vibrant digital citizenry—engaged, resourceful, and resilient. The field was largely unified in its optimism. More information meant better awareness. More voices meant richer perspectives. More sharing meant stronger communities. Emergency managers were encouraged to embrace social listening, to treat platforms as early-warning systems, and to trust that the crowd, on balance, was wise. Researchers focused on identifying credible voices, filtering out rumors, and understanding the social dynamics that made some communities exceptionally effective. This optimism shaped the field’s questions, its methods, and its ethical stance. What was rarely questioned was the underlying framework itself: that social media in crises were essentially positive forces, extensions of the public sphere that could and should be harnessed for the collective good.

The pandemic shattered that optimism with brutal efficiency. As COVID-19 spread across the globe, a parallel epidemic of information moved alongside it—an “infodemic,” as the World Health Organization memorably called it. The information environment was not merely polluted; it was weaponized. Conspiracy theories about engineered viruses, fabricated miracle cures, and malicious claims about vaccine dangers multiplied at digital speed. Every hour brought new falsehoods, some harmless, many deadly. A viral post recommending bleach as a treatment wasn’t an abstract concern; it was a public health catastrophe waiting to happen. Another claim, linking 5G towers to the spread of the virus, led to arson attacks on telecommunications infrastructure in multiple countries, physically severing communication lines that hospitals and emergency services depended on. For crisis informatics researchers, the ground shifted beneath their feet. The discipline that had spent years helping emergency managers find useful needles of information in haystacks of noise suddenly found itself confronting a crisis where the noise itself was the enemy. The stakes were life and death, and the enemy was not the storm or the earthquake—it was the platform itself, algorithmically amplifying dangerous nonsense alongside life-saving guidance. Policymakers, who had previously treated social media as a peripheral concern, now recognized these platforms as decisive battlegrounds for public health. They wanted tools for detecting misinformation, methods for countering its spread, and insights into why people believed it. Crisis informatics scholars, whose expertise in rumor detection and source validation suddenly appeared exquisitely relevant, found themselves thrust into urgent, high-stakes conversations about governance, regulation, and the limits of free speech. The questions that emerged were profound: How do we distinguish between harmful misinformation and merely unpopular opinions? Who has the authority to make that call? And what happens when the platforms themselves are complicit in amplifying the very dangers they claim to police?

Just as these demands were escalating, the research environment itself began to disintegrate. The most dramatic rupture came with the transformation of Twitter into X. Under new ownership, the platform that had been the lifeblood of crisis informatics research—the source of vast data streams that scholars had used for over a decade to track emerging narratives, map information flows, and identify influential actors—became dramatically less accessible. The academic API, which had provided researchers with the “firehose” of real-time tweets, was walled off behind a paywall so steep that most academic institutions simply could not afford it. For many scholars, overnight, the data they had built their careers and their methods upon vanished. But the challenges went far beyond access. The platform’s moderation policies were radically transformed, overturning assumptions that researchers had long taken for granted. Fact-checking labels were removed, previously suspended accounts were reinstated, and content moderation was scaled back in the name of an absolutist vision of free expression. The coordinated efforts of public health organizations to flag vaccine misinformation were dismantled. For crisis informatics, the implications were staggering. Researchers could no longer systematically track how misinformation narratives evolved over time, because the data was gone and the platform’s aggressive amplification of controversial content had warped the landscape beyond recognition. The historical record of social media during COVID-19—arguably the most consequential crisis of the digital age—became fragmented, incomplete, and potentially lost to future scholarship. The study under discussion here frames this as more than an inconvenience. It represents a fundamental methodological crisis. The very platforms that had become the primary arenas for public emergency communication were now less transparent, less accessible, and more unpredictable than at any point in the field’s history. Crisis informatics, a discipline built on the promise of studying these platforms, is now forced to reconsider its foundations entirely.

To make sense of this upheaval, the study introduces the concept of ’emergency politics.’ This is the insight that states of emergency—pandemics, natural disasters, terrorist threats—routinely trigger exceptional governance measures that reshape the relationship between citizens, states, and media institutions. During COVID-19, governments around the world invoked emergency powers with varying degrees of justification and restraint. Some took down posts they deemed dangerous, often under vague and hasty “fake news” laws. Others pressured platforms directly to suppress content, presenting it as legitimate public health intervention. The authors argue that crisis informatics has been far too slow to engage with this political dimension. The field, they contend, has treated social media primarily as neutral resources—as tools that citizens use to inform and organize—while paying insufficient attention to how emergency measures actively structure, constrain, and politicize the flow of information. Was the suppression of a post questioning vaccine efficacy a legitimate public health measure or an overreach of state power? Were platforms acting in good faith when they amplified certain voices and silenced others? The study insists that these questions can no longer be ignored. The emergency politics framework reveals a deeper truth: social media are not neutral channels. They are contested terrains where the boundaries of public speech, state power, and corporate control are constantly being negotiated. In ordinary times, these tensions remain largely hidden. But during emergencies, they become starkly visible. The pandemic brought these tensions to the surface, forcing crisis informatics to confront the uncomfortable possibility that the platforms enabling citizen coordination could simultaneously enable surveillance, censorship, and the silencing of legitimate dissent. The field’s romantic attachment to digital citizen engagement, the authors argue, must be replaced by a more critical and reflexive stance.

The study’s contributions reach beyond crisis informatics into the broader scholarly conversation about surveillance, freedom of expression, and democratic governance during emergencies. It acknowledges that some emergency measures may be necessary and even beneficial—swift removal of dangerous misinformation can indeed save lives. But it insists that such measures cannot be evaluated in isolation. Who decides what constitutes dangerous information? Whose voices are silenced, and whose are amplified? What happens to the historical record of dissent when platforms purge content at the request of governments? These questions, long central to political communication scholars, have rarely been integrated into crisis informatics research. The study calls for a more interdisciplinary future—one in which crisis informatics scholars are as fluent in the politics of surveillance as they are in algorithmic methods. Looking forward, the authors articulate a new research agenda for a post-pandemic world. It begins with developing methods that can operate without privileged access to platform data, perhaps through federated networks or archived collections. It demands deeper theoretical engagement with platform governance and its political consequences. And it requires sustained critical analysis of how emergency politics reshapes the digital public sphere. The pandemic, the authors remind us, has not ended. It has evolved, and the information environment around it has evolved with it. For crisis informatics to maintain its relevance, it must shed its innocence—the comfortable assumption that social media are uncomplicated tools for good—and embrace the messy, contested, political reality of these platforms. Only then can the field fulfill its true promise: helping us understand not just how information flows during a crisis, but who is hurt, who is helped, and who gets to decide the difference.