Large language models (LLMs) can summarize, aggregate, and deliver localized climate-related data efficiently and cost-effectively. The authors present ClimSight, a simple prototype that demonstrates how LLMs can provide actionable climate information to anyone with access to a computer. ClimSight combines LLMs with geographical and climate data to address the challenge of obtaining user-focused climate information at the local scale. The system uses climate model simulations and geographical data to evaluate the impact of climate change on specific activities, such as agriculture or wind turbine installation. ClimSight is designed to answer user queries about climate change impacts, such as whether a particular crop can be grown in a specific location. It uses data from climate models and geographical information to provide insights into temperature, precipitation, and wind patterns. The system is currently limited in scope, using low-resolution climate data and not accounting for uncertainties. However, it shows promise in summarizing and communicating climate information effectively. The authors suggest that expanding ClimSight by incorporating high-resolution data and advanced climate models could improve its capabilities. They also highlight the importance of ensuring the reliability of the information provided, as LLMs can sometimes generate inaccurate or misleading information. Additionally, integrating trusted sources of information, such as scientific publications and local regulations, could enhance the system's accuracy and usefulness. ClimSight has the potential to democratize climate information by making it accessible to a broader audience. It can support local decision-making by providing location-specific guidance, overcoming traditional barriers such as scalability, expertise, and geographical constraints. The system is cost-effective, with each query costing approximately 6 cents using the OpenAI GPT-4 API. The authors also suggest that transitioning to open-source models like LLAMA2 could further democratize the development and use of ClimSight.Large language models (LLMs) can summarize, aggregate, and deliver localized climate-related data efficiently and cost-effectively. The authors present ClimSight, a simple prototype that demonstrates how LLMs can provide actionable climate information to anyone with access to a computer. ClimSight combines LLMs with geographical and climate data to address the challenge of obtaining user-focused climate information at the local scale. The system uses climate model simulations and geographical data to evaluate the impact of climate change on specific activities, such as agriculture or wind turbine installation. ClimSight is designed to answer user queries about climate change impacts, such as whether a particular crop can be grown in a specific location. It uses data from climate models and geographical information to provide insights into temperature, precipitation, and wind patterns. The system is currently limited in scope, using low-resolution climate data and not accounting for uncertainties. However, it shows promise in summarizing and communicating climate information effectively. The authors suggest that expanding ClimSight by incorporating high-resolution data and advanced climate models could improve its capabilities. They also highlight the importance of ensuring the reliability of the information provided, as LLMs can sometimes generate inaccurate or misleading information. Additionally, integrating trusted sources of information, such as scientific publications and local regulations, could enhance the system's accuracy and usefulness. ClimSight has the potential to democratize climate information by making it accessible to a broader audience. It can support local decision-making by providing location-specific guidance, overcoming traditional barriers such as scalability, expertise, and geographical constraints. The system is cost-effective, with each query costing approximately 6 cents using the OpenAI GPT-4 API. The authors also suggest that transitioning to open-source models like LLAMA2 could further democratize the development and use of ClimSight.