AI Analytics Enhances Policymaking on Climate Risk Perceptions

A groundbreaking study published in Sustainability reveals that leveraging artificial intelligence (AI) can significantly enhance our understanding of public perceptions regarding climate risks. The approach centers around AI-driven analytics of social media platforms, particularly X (formerly Twitter), which is increasingly a venue for public discourse on environmental issues. By analyzing a staggering 29,576 English-language posts collected in December 2025, researchers demonstrated that AI tools could help track interpretations of climate risks, identify the spread of misinformation, and highlight effective messaging for climate adaptation and sustainability initiatives.

The sentiment analysis revealed a predominantly negative tone in climate-related discussions online, with 18,654 posts categorized as negative, in stark contrast to just 6,550 positive posts and 4,372 neutral contributions. The advanced deep learning model utilized in the study achieved an impressive 92.5% accuracy rate in sentiment detection, showcasing its reliability in categorizing the emotional undertones of posts. This analysis revealed a public conversation that is heavily dominated by sentiments of fear, urgency, and discontent, which presents a significant challenge for policymakers. The dominate narrative reflects how discussions about global warming are predominantly framed as a crisis, instead of being viewed as a long-term environmental concern.

The content analysis showed that users were particularly vocal about their anxieties surrounding extreme weather events and their frustration with perceived governmental inaction on climate issues. Phrases indicating concern about the future of the planet and critiques of human-induced environmental damage were common. This duality in climate conversation—between genuine environmental concern and skepticism—indicates a complex landscape of public sentiment, further complicating the efforts of officials and communicators tasked with promoting climate action.

Moreover, the study identified two overlapping discourse forms within the negative posts. The first group reflects genuine worries regarding ecological instability, while the second encompasses a sense of skepticism toward the scientific consensus, where users evoke references to cold weather and frequent snowfalls in a bid to challenge the legitimacy of climate phenomena. This intricate blend of anxiety and skepticism complicates the public narrative and necessitates strategic communication methods by climate advocates and policymakers.

In contrast, the analysis of positive posts, albeit fewer in number, indicated an emerging narrative that emphasizes awareness and a call to action. Messages that focus on scientific advocacy, clean energy initiatives, and shared responsibility highlighted the necessity of proactive engagement from the public in combating climate challenges. Rather than denying the significance of global warming, this positive discourse highlights pathways toward resilience and adaptive measures.

The findings underscore the potential value of constructive communication strategies that could pivot the prevailing public apprehension into active community engagement on climate issues. Given this shift in narrative, policymakers are encouraged to capitalize on the insights derived from this AI-driven analysis to tailor their messaging effectively and foster a more engaged citizenry around sustainability efforts.

Furthermore, this study illustrates the critical role AI can play in shaping climate communication approaches, guiding decision-makers to better understand public sentiment dynamics and, ultimately, enhancing policy effectiveness in the face of climate change.

The implications of these findings are substantial for the Department of Energy, EPA, and other relevant agencies tasked with addressing climate issues. By focusing on the emotional drivers of climate change discourse as revealed by AI analytics, these agencies can develop targeted strategies that not only inform but also mobilize the public toward informed climate action.

• AI analysis of 29,576 posts reveals public sentiment on climate change.
• 18,654 posts labeled as negative; 6,550 positive posts noted.
• Study achieved 92.5% accuracy in sentiment detection.
• Climate discourse heavily influenced by themes of fear and urgency.
• AI tools can help policymakers track climate risk perceptions.
• Positive messaging emphasizes solutions and collective responsibility.
• Skepticism evident in discussions referencing cold weather to question climate data.
• Effective communication is essential for transforming concern into engagement.
• Findings offer actionable insights for agencies like EPA and DOE.