By Tom Ellison, Erin Sikorsky, and Noah Fritzhand

Information manipulation is increasingly recognized as a contributor to climate risks. 2025 saw landmark academic assessments on the topic, as well as the first action on climate information integrity at a UN climate summit. Meanwhile, bad actors take advantage of opportunities to propagandize, sow confusion, and undermine trust as the impacts of climate change intensify and the stakes of policy action grow. These mutually reinforcing challenges jeopardize security and democracy, especially amid volatile geopolitics, rapid change in the technology and media landscape, and US federal reversals on climate policy and information integrity. 

This raises questions for a range of US actors amid intensifying extreme weather. How can state and local officials build resilience and respond to emergencies when facing an unsupportive federal government and global, minimally regulated information threats? What are the implications for US military disaster relief and readiness when information manipulation threatens political cohesion and civilian communities? How can journalism or tech policy serve climate security by mitigating mis/disinformation? And how can academia, civil society, and community groups better collaborate to exchange information and expertise? 

The Center for Climate and Security (CCS) recently completed a foresight exercise in Florida to explore these questions. Co-hosted by the University of Miami’s Climate Resilience Institute, the event brought together a diverse mix of expertise, including local resilience and emergency management, national security and foreign policy, communications and information integrity, social and natural sciences, and local climate education and activism. Participants heard from senior homeland security and defense speakers, then engaged in a facilitated scenario exercise exploring a plausible extreme weather, information, and national security crisis in Florida. The discussion highlighted several key themes. 

A Plausible Future Underscores Risks

Participants considered a hypothetical scenesetter imagining intersecting crises in 2030, briefly summarized below.

Florida faces intensifying heat, storms, and flooding, contributing to a housing cost and insurance crisis that has birthed a contentious campaign for a ballot initiative on housing affordability. A recent hurricane has caused widespread damage in Tampa Bay that the city is still recovering from. Russia extracted a settlement in Ukraine and is acting aggressively in the Baltics, while Cuba faces protests over an economic crisis and blackouts caused by a recent hurricane. The information ecosystem remains characterized by siloed social media discourse, a struggling journalism sector, and continued rapid LLM/chatbot development. With mid-term elections approaching that will decide the housing affordability proposal and possibly control of Congress, another intense hurricane devastates Cuba and Haiti and is landing in Miami. Civilian government agencies, the US military, Florida populations, online communities, and Russian, Cuban, and extremist propagandists begin responding.

Participating experts considered the hypothetical crisis largely plausible (or in some instances overly cautious), spotlighting the climate, security, and mis/disinformation challenges on the current trajectory. 

Such a crisis poses risks for key actors. For example, the military’s US Southern Command would be challenged to implement up to four contingency plans simultaneously (securing its own headquarters in Miami, mobilizing humanitarian assistance/disaster relief in Haiti and Cuba, and/or monitoring for potential Cuban instability), possibly taxing military assets in high global demand, such as reconnaissance drones. The National Guard would also be called upon. Already limited government public health, financial, and other resources would be strained in the event of successive hurricanes–a plausible but so–far unseen scenario for Florida. Meanwhile, Florida communities would have to navigate mis- and disinformation that could endanger populations, hamper US military or other critical functions, further polarize society, and erode trust in the validity of upcoming voting. And actors like Russia, Cuba, domestic extremists, and online fraudsters would have opportunities and incentives to exploit the situation–hindering US relief and recovery, victimizing US populations, harming US security interests, and undermining democracy. 

Policy Leadership, Generative AI’s Role, and Other Wildcards

Several factors could shape the trajectory of such a crisis in better, worse, or different ways for Florida communities and national security, with varying degrees of uncertainty by 2030-2031. 

Discussions focused on two variables. First was whether the locus of 2030 policy action is more top-down (with major roles for the US federal government, Florida elites, and corporations) or bottom-up (with the center of gravity being local government and community networks). The role of federal or state government top-down action cannot be easily replaced, in terms of providing funding and legal authorities, or at the very least a politically permissive environment for local or nongovernmental action. At the same time, locally led and decentralized actions can be more flexible and responsive to community needs, even if under-resourced or fragmented. In reality, a combination is needed.

A second key variable was the harmful or helpful role of more widespread generative AI, namely chatbots and synthetic media. Such technologies pose multifaceted risks and potential opportunities in climate security and information crises. Amid minimal regulation and transparency, chatbot training data has already been a target for Russian influence efforts. The ideological manipulation by xAI CEO Elon Musk has highlighted the risks of further concentrating control of public information with unaccountable technology companies. Experts also worry that ubiquitous synthetic media could broadly erode public willingness and ability to discern fact from fiction. However, if properly applied and regulated, generative AI and machine learning could also contribute to climate resilience and information integrity. For example, such tools could rapidly and automatically disseminate multilingual emergency information to individuals, detect problematic social media content for users, aid in resilience planning, and speed insurance payouts. 

Participants discussed other key factors that will shape the 2030 policy future beyond these two variables, including:

• The track record of US disaster responses between 2025 and 2030, which will affect whether or not responses by 2030 enjoy sufficient resources and political legitimacy, amid cuts to FEMA, NOAA, and the social safety net. 
• The resilience of Florida’s power, water, nuclear energy, and other infrastructure, which will determine the likelihood of disruptions that could hamstring military and community responses and fuel mis/disinformation. 
• The perception of US military activity in Latin America by 2030 and how it shapes the reaction to US military hurricane relief or contingency preparations around Cuba.
• The state of US media independence and local journalistic capacity, and its implications for whether Florida communities have access to trustworthy information and government accountability as a bulwark against mis/disinformation. 

Takeaways and Opportunities

These factors combine to imply a range of potential outcomes to the scenario. In a worst-case scenario, mutually reinforcing climate-related impacts, information manipulation, and security challenges could collectively contribute to catastrophic harms to Florida communities, democratic trust, and US national security. However, prudent action across subnational and federal government levels, academia, civil society, media, and the private sector could help avoid the most negative outcomes, and such challenges could force innovations and new coalitions that pave the way for future resilience building. 

Regardless,  any scenario at the nexus of extreme weather, information, and security will bring new questions and challenges. For example, even the most healthy AI futures could introduce technological dependencies that create new vulnerabilities and exacerbate marginalization of those lacking digital access. Even well-resourced, whole-of-society efforts would have to overcome coordination and prioritization challenges (e.g., between civilian and military; between local, state, and federal; between government, civil society, and academia; etc.). 

With that in mind, the group emphasized several recommendations and takeaways: 

• Information exchange with communities: Climate, housing, public health, and other organizations working directly with key populations are best positioned to understand the information diets and concerns of communities, including via existing “train-the-trainer” programming. These channels can be valuable for both building resilience to information manipulation and detecting problematic narratives early. With better organization and trust-building, more systematic information-sharing and training in all directions between local government, communities, and civil society can help mitigate both community harms and related security risks.
• Horizontal collaboration: Sharing information, resources, and best practices on security and disinformation threats during extreme weather across state and municipal lines is key, especially if federal support is missing. Existing coalitions and associations can be better networked and supported on these topics, such as Florida’s South Florida Climate Compact. Similar venues exist in other states, nationally (e.g., America’s All In or ICLEI USA), and for key professions (e.g., resilience officer networks, the National Emergency Management Association, or cybersecurity fora). US states and cities can also absorb lessons learned by communities overseas facing similar climate hazards or governance challenges. 
• Early partnerships, “prebunking,” and nontraditional voices: The voices most capable of mitigating mis/disinformation during disasters or key climate debates are often outside official channels, such as influencers, athletes, youth activists, faith leaders, or meteorologists. These voices are critical when disasters strike, and their ability to boost credible information, maintain social cohesion, and encourage democratic governance brings significant benefits to human and national security. It is crucial to develop trusted relationships and mis/disinformation literacy with such voices well before disasters strike (sometimes described as “prebunking”), after which it is too late. 
• Unique functions: While local and nongovernmental efforts are important, certain actions lie only with state, national, and/or international government authority. Examples include certain technology regulation and standard-setting, funding and coordinating nationwide disaster relief and science (e.g., FEMA and NOAA), and administering elections. Work on climate, security, and mis/disinformation challenges must address multiple scales and jurisdictions with relevant recommendations.

Conclusion & Next Steps

The exercise in Florida highlighted the urgency of climate security and misinformation risks, opportunities for action in the current environment, and follow-up lines of inquiry. The event was the first exercise in a larger CCS project supported by the Knight Foundation on climate security, mis/disinformation, and democratic governance in the United States. The project will produce additional research, practitioner resources, and engagements at Stanford University and New York Climate Week over the coming year, where CCS will build on these insights.