Weathering Extremes: New Toolkit Tackles Disinformation in an Age of Climate-Fueled Disasters
From the heatwaves that scorched Spain and Portugal to the wildfires that ravaged parts of the United States, several of this summer’s extreme weather events were accompanied by rampant climate disinformation, even as organizations like the European Digital Media Observatory sounded alarms about the predictable spread of false narratives during disasters. As extreme weather events intensify, the information landscape around them is also becoming more complex and chaotic. The rapid growth of Large Language Models (LLMs) and the ubiquity of social media have created new avenues for users to generate, consume, and spread false information, making it easier than ever for bad actors to exploit crises. Against that backdrop, understanding and responding to information threats amid disasters has become both more urgent and more challenging. A new toolkit, Weathering Extremes: A Toolkit for Information Resilience Amid Disaster, aims to meet that challenge head-on. Released earlier this month by the Center for Climate and Security, the toolkit was authored by Jennie King, Non-Resident Fellow at the Center for Climate and Security, and provides a deep dive into the most salient information threats associated with extreme weather events. It offers practical guidance on how to assess and address those risks before, during, and after disasters. Crucially, the toolkit frames information resilience — defined as the ability to resist and recover from information threats — as “a key element of disaster preparedness, response, and recovery.” Rather than treating false narratives as an afterthought, the report argues that safeguarding the information environment must be integrated into every phase of disaster management, from planning and readiness to emergency response and long-term rebuilding. The timing is apt: this summer’s headlines have repeatedly shown that disinformation can spread as quickly as floodwaters, with potentially deadly consequences for communities already struggling to survive.
Extreme weather is a particularly fertile ground for disinformation, according to contributing author Tom Ellison, Deputy Director of the Center for Climate and Security. During the report launch, Ellison noted that extreme weather events are a flashpoint for several reasons. First, they are high-stakes: misconceptions or confusion about threats can cost lives. When people do not know whether to evacuate, where to seek shelter, or whether official warnings are trustworthy, the consequences can be catastrophic. Second, because these events tend to be high-profile, they attract malign actors who weaponize crises for “financial, political, or personal gain.” Disasters are high-visibility moments that drive many people online in search of live updates, and this provides a unique opportunity to amplify and launder ideas to a mainstream audience. As King writes in the toolkit, “Extreme weather events now attract authoritarian states, political extremists, vested interests, and digital profiteers alike.” The emotionally fraught environment created by extreme weather also makes people more vulnerable to manipulation. Fear, uncertainty, and desperation can override critical thinking, making individuals more likely to share alarming content without verifying its accuracy. This toxic combination of high stakes, high visibility, and heightened emotional vulnerability makes extreme weather events an ideal breeding ground for disinformation campaigns, whether they are launched by foreign adversaries seeking to sow chaos, domestic political actors looking to score points, or opportunistic scammers hoping to profit from confusion and fear.
The short- and long-term impacts of this information chaos are significant and well documented. In the wake of Hurricanes Helene and Milton in 2024, for example, social media accounts spread conspiracy theories about government control of the weather and circulated false claims that FEMA was diverting funds to foreign aid and asylum seekers instead of helping storm victims. Investigations later revealed that Kremlin-backed media networks and other Russian-linked sources were “seeding this narrative” across platforms and elevating domestic disinformation. The consequences were not limited to online confusion. Public officials faced death threats, and safety concerns forced some emergency workers to evacuate their posts, hampering response efforts at a critical moment. This illustrates how information threats can translate directly into physical threats, endangering the very people tasked with keeping communities safe. The toolkit warns that in the long term, information chaos amid disasters can erode the public’s confidence in the information they receive, undermining broader governance and security. When people lose trust in official sources, they become more susceptible to false narratives in future emergencies, creating a vicious cycle of distrust and vulnerability. Moreover, disinformation can exacerbate existing social divisions, fuel conspiracy theories, and make it harder for communities to recover and rebuild after a disaster. In this sense, information resilience is not merely a communications issue but a core component of national and homeland security, as well as public safety.
Among the toolkit’s many recommendations, its guidance on Large Language Models stands out as the most distinctive and forward-looking. LLMs have undoubtedly changed the landscape of mis- and disinformation, from information generation to uptake, with people increasingly relying on LLMs as everyday sources of information. In fact, the toolkit cites research by Semrush that finds that AI searches could surpass Google traffic by the end of 2027. This shift has profound implications for how information is created, consumed, and trusted. LLMs have enabled malign actors to produce and share misleading or false information at a larger scale, more cheaply, and more effectively than ever before. For instance, there are high-profile cases of bad actors using “LLM seeding” tactics to make disinformation more visible to — and more often cited by — chatbots. If users do not verify the sources or accuracy of chatbot responses, false narratives can quickly gain legitimacy in their minds. To stay relevant and visible in this new information environment, King writes that “the same tactics must be employed to elevate high-trust information on an ongoing and dynamic basis.” The toolkit recommends that organizations seeking to counter mis- and disinformation consider using generative engine optimization (GEO) strategies that can make their content more prominent in chatbot responses and AI overviews. Examples identified in the toolkit include building visibility on platforms like Reddit and YouTube, adjusting the format of website content to be “AI readable,” and repeating information on “higher-trust sites” like traditional mass media outlets and .gov domains. These strategies are designed to ensure that when people turn to AI tools for answers during a disaster, they encounter accurate, authoritative information rather than conspiracy theories and propaganda.
Identifying the right actors and providing them with tools, however, is just one step among many. The toolkit is designed primarily for state, local, and tribal government officials, including emergency managers, public information officers, and resilience officers, who oversee disaster preparedness and response. But it also encompasses broader stakeholders — such as utilities, healthcare providers, and first responders — and their role in helping assess readiness, build trust with communities, and respond to information threats. These actors are on the front lines of disasters and are often the first to face questions from the public. Yet they frequently lack the resources and technical capacity to implement sophisticated counter-disinformation measures. State and local resilience offices are often understaffed and working with limited technical capacity, and they may struggle to implement generative engine optimization or track their own visibility across platforms they do not control. Moreover, GEO is new enough as a practice that additional evidence is needed to understand how it can reliably elevate high-trust information over disinformation in real-world crisis scenarios, as well as how AI visibility can be effectively evaluated. Despite these challenges, the toolkit argues that proactive investment in information resilience is essential. By giving these actors the tools and avenues to build trust in reliable information and guard against manipulation, the toolkit aims to “shift the balance away from crisis comms and fact-checking towards upstream engagement and risk mitigation.” In other words, instead of waiting until false narratives have already spread and caused harm, officials should work continuously to strengthen the information environment so that accurate, trustworthy content is more visible and more resilient to attack.
Ultimately, addressing the challenge of mis- and disinformation amid extreme weather events is about building systems that not only ward against the proliferation of false information but also equip users with the tools to question and resist it. Weathering Extremes provides a framework for doing exactly that, urging officials and communities to treat information resilience as a core element of disaster preparedness, response, and recovery. In a world where AI-generated content and social media can amplify falsehoods within minutes, proactive, upstream engagement is no longer optional. The toolkit’s emphasis on generative engine optimization, partnership with trusted local actors, and long-term trust-building reflects a growing recognition that the information environment is as critical to disaster outcomes as physical infrastructure. Just as communities build levees, reinforce power grids, and stockpile emergency supplies, they must also invest in the social and digital infrastructure that helps people find reliable information when they need it most. The stakes could not be higher: in an era of intensifying climate change, extreme weather events will only become more frequent and severe, and so too will the information threats that accompany them. The toolkit makes clear that information resilience is not a luxury or an afterthought but a vital part of keeping communities safe. By adopting the strategies outlined in Weathering Extremes, emergency managers, public officials, and community leaders can help ensure that when the next disaster strikes, the truth is loud, clear, and impossible to drown out.