Brief Author Note

Did… did we win? 😟 So, if you have been keeping track of the news, the COP30 conference just wrapped up last week to a surprisingly disappointing result depending on your perspective.

Here are a couple of takeaways from the conference from some news articles that I found:

• COP30 falls far short on climate justice

• What COP30 did and what COP30 did not do

• AI is guzzling energy for slop content – could it be reimagined to help the climate?

• Artificial intelligence sparks debate at COP30 climate talks in Brazil

• Another Cop wrecked by fossil fuel interests and our leaders’ cowardice – but there is another way

• Aramco believes AI can help double oil-well productivity, hydrocarbons here to stay

The Sparknotes version is that AI was front and center as a solution to the climate crisis BUT there was little focus on the environmental impacts of AI or data centers apparently on communities and areas. In addition, petronations and the fossil fuel industry successfully fought against fossil fuel phaseouts in the COP30 text known as the Global Mutirão or Global Collective Effort. So… basically climate denial?

Watt you later! ⚡️ 

Nate

Quick Links!

• Podcast Episodes

• Scientific Papers

• Policy Documents

• Multimedia

• Organizations

• Tools

• Fellowships Corner

• Submit Resources

COP30 Covered: António Guterres’s Call to Action and AI’s Climate Impact (AI & Policy/ESG): This podcast episode breaks down the UN Secretary-General’s urgent message at COP30 and examines how AI is being positioned as both a powerful climate solution and an emerging environmental risk.

Digital Platforms, AI, and the Climate Information Environment (AI & Environment): This podcast explores how digital platforms and AI systems shape what people see, search, and share about climate, highlighting how algorithmic design fuels misinformation, greenwashing, and visibility biases while also carrying significant energy and resource demands.

Climate, AI, and Me? (AI & Environment): This podcast examines how AI tools are increasingly embedded in everyday software and argues that their growing energy and water demands mirror the broader climate crisis, raising questions about whether the benefits of generative AI outweigh their environmental footprint.

Investigating Awareness and Behavioural Intentions From AI Water Footprint Digital Posters in Higher Education Settings (AI & Water Resources): This study tests whether digital posters at the University of Glasgow can raise awareness of AI’s hidden water footprint, finding that many students were surprised by AI’s water use and that exposure increased their intention to think critically before using AI tools. The results show that low-cost, visually engaging interventions can meaningfully shift sustainability awareness and behaviour on campus.

Intelligence Per Watt: Measuring Intelligence Efficiency of Local AI (AI & Energy): This paper evaluates whether small, locally executed language models can handle real-world LLM queries on power-constrained devices, introducing “intelligence per watt” (IPW) as a unified metric for capability and efficiency. The findings show rapid progress—from 23.2 percent to 71.3 percent real-world query coverage between 2023 and 2025—demonstrating that local LMs can meaningfully reduce reliance on centralized cloud infrastructure.

ScaleONet: Scalable and Control-Oriented Modeling of Building Cluster Thermal Dynamics Using Deep Operator Networks (AI & Energy): This study introduces ScaleONet, a deep operator network that models thermal dynamics across large, heterogeneous building clusters far more efficiently than conventional simulators or standard data-driven models. Validated across thirty Belgian buildings, the system achieves sharply lower prediction error with millisecond-level inference, enabling district-scale energy flexibility, real-time control, and scalable optimization.

AI-Powered Microgrids: Optimizing the Balance — The Path Toward Coupling Economic and Environmental Benefits (AI & Energy): This report explains how AI-enabled microgrids can optimize energy generation, storage, and consumption to reduce costs while improving resilience and lowering carbon emissions. It highlights how real-time forecasting, load management, and distributed energy coordination unlock both economic and environmental benefits at scale.

Climate Change: What Is Computing’s Responsibility? (AI & Sustainability): This manifesto outlines the computing sector’s ethical and environmental responsibilities in a warming world, arguing that researchers, developers, and institutions must confront the energy, material, and social impacts of digital technologies. It calls for systemic accountability, sustainability-focused research agendas, and a shift toward climate-aligned computing practices.

Update on Scientific Findings on the Interactions Between Agriculture, Food Systems and Climate Change (AI & Agriculture): This FAO report synthesizes the latest science on how agriculture and food systems both drive and are impacted by climate change, emphasizing the need for integrated mitigation, adaptation, and resilience strategies. It highlights evidence on emissions trends, land-use pressures, and policy interventions necessary to support sustainable and food-secure futures.

Protecting Colombia’s Endangered Wildlife with Project SPARROW (AI & Biodiversity & Conservation): This short documentary highlights Project SPARROW, an AI-powered conservation initiative developed by Microsoft’s AI for Good Lab to protect Colombia’s endangered Middle Magdalena Valley. Using solar-powered cameras and acoustic sensors that stream data via satellite, AI models analyze wildlife activity in real time, giving researchers instant ecological insights and offering a scalable, open-source template for global biodiversity protection.

AI for a Planet Under Pressure Panel (AI & Multiple Applications): This panel examines whether AI can be used responsibly and effectively to tackle complex sustainability challenges, drawing on findings from the report AI for a Planet Under Pressure. Speakers explore AI’s research value across eight issue areas while assessing the risks, limitations, and opportunities of applying advanced models to climate, ecological, and social systems.

How to Turn AI Into a Power Plant with Arushi Sharma Frank (AI & Energy): In this episode of Power Perspectives, Arushi Sharma Frank explains how AI data centers can shift from grid stressors to grid assets through orchestration, flexible load management, and smart market design. Drawing from her experience across Tesla, Exelon, and ERCOT engagement, she outlines how advanced AI and energy policy can reshape dispatchability and the future of power systems.

Vivent Biosignals (AI & Agriculture): Based in Europe, Vivent Biosignals uses advanced biosensors and AI algorithms to decode plant signaling in real time, detecting crop stress and environmental responses long before visual symptoms appear. Their technology helps farmers improve yields, reduce inputs, and strengthen sustainable agriculture across the region.

Plentify (AI & Energy): Based in North America, Plentify builds AI-enabled home energy solutions that coordinate across entire neighborhoods to boost grid resilience while reducing household energy consumption. By combining hardware, intelligent control systems, and user-centric design, they support more reliable and sustainable energy systems for communities.

Endolith (AI & Circular Economy): Located in North America, Endolith integrates microbes, synthetic biology, and cloud-based analytics to develop more sustainable copper extraction methods. Their approach reduces environmental impact and sets a new standard for circular, resource-efficient mining practices.

Sparrow Client (AI & Biodiversity/Ecology): Developed by Microsoft’s AI for Good Lab in the United States, Sparrow Client is an open-source, AI-powered edge computing tool designed to monitor and protect wildlife in remote and hard-to-reach ecosystems. By enabling real-time analysis of audio and image data, it helps conservation teams detect species activity and ecological changes without needing on-site presence (FREE!)

Sustainable Games Standard (AI & Sustainability): Created by the Sustainable Games Alliance, the Sustainable Games Standard provides developers, sustainability teams, and service providers worldwide with actionable guidance to reduce emissions and embed sustainability across the gaming lifecycle. The framework offers activity-based insights to help studios optimize operations, improve performance, and position gaming as the most sustainable major entertainment sector. (FREE!)

Digital Sustainability Resources (AI & Sustainability): This interactive sustainability app helps organizations benchmark their web sustainability maturity, offering tools and reference frameworks to assess digital footprints and identify improvement pathways. Designed for global use, it supports teams looking to reduce energy and emissions across websites and digital assets (FREE!)

Cosmos Fellowship with Oxford’s Human-Centered AI Lab (Open globally to experienced researchers & junior talent): This fellowship supports individuals with world-class AI potential and deep philosophical insight, focusing on “philosophy-to-code” projects that turn concepts like autonomy, virtue, decentralization, and inquiry into practical AI system design. Fellows work with Oxford’s HAI Lab and the Cosmos Institute’s 400+ member network, gaining mentorship, exclusive seminars, and an environment suited for producing impactful research, software, and publications. The program lasts up to one year (with possible extension), offers USD $75,000 per year plus travel support, and can be completed in Oxford or remote/hybrid. Applications close December 6, 2025.

AI Policy Leadership Network (Open to mid- to senior-level professionals, U.S.-based): This part-time program brings together about 25 experienced professionals from across the U.S. AI policy ecosystem including government, think tanks, industry, the military, and the intelligence community to build relationships and engage directly with leading figures shaping national AI strategy. The program runs from mid-March to late July 2026 and includes monthly evening sessions in Washington, DC and a fully funded, multi-day learning trip to the San Francisco Bay Area. Designed to fit alongside demanding full-time roles, the program requires 6–25 years of professional experience (with at least four in policy-relevant organizations) and a current or emerging role connected to AI policy. Applications close December 18, 2025.

Tarbell Fellowship (Open globally to early and mid-career journalists OR individuals with a strong AI background interested in journalism):
This year-long program (June 2026–May 2027) supports early- and mid-career journalists who want to specialize in covering artificial intelligence. Fellows receive a nine-month placement at a major newsroom, participate in a 10-week study group on AI and journalism fundamentals, and attend a weeklong journalism summit in the San Francisco Bay Area. The fellowship provides $60,000–$80,000 stipends depending on location, with $90,000–$110,000 for senior fellows (five or more years of experience). Open to applicants worldwide, with visa support where possible. Applications close January 7, 2026

BONUS: You can check out Alisar Mustafa’s November roundup of AI/Tech Policy fellowships. 😀 

If you have resources that you wish to submit to the AI & Environment Resource Hub, fill out the Google Form! Thank you for suggesting resources to add to the Resource Hub. 🙃 

That’s it for this week.

Thanks for reading the 28th issue of The Climate Code! It means a lot to me and we have more coming in the future, so definitely stick around! 💚 

Nate

P.S.

New here? Check out the AI & Environment Resource Hub if you have no idea what it is. 😆