How AI can Improve Energy Use

AI has no place in sustainability. There’s a familiar stance I hear a lot in sustainability circles. AI uses a lot of energy. So using it for sustainability sounds… contradictory. But that argument misses the bigger picture. AI isn’t just consuming energy. It’s helping us use less of it too. Used well, AI is already solving real sustainability problems. Not hypotheticals. Not R&D lab demos. Live, operational tools that help businesses reduce emissions, speed up reporting, and make better decisions. Here’s what that looks like in practice: 1. Energy grid optimisation In the UK, the National Grid is using AI to forecast solar energy production by analysing satellite images and weather data. If clouds are expected to lower solar output in, say, Cornwall 30 minutes from now, the grid can prep alternative sources in advance. That means fewer blackouts and lower emissions from fossil backup plants. DeepMind did something similar for wind power. Their AI predicted wind farm output 36 hours in advance, which increased the commercial value of wind energy by around 20 percent. Why? Because energy providers could schedule when to send power to the grid with more certainty. 2. Streamlined carbon accounting AI tools now scan invoices, utility bills and PDF reports to pull out emissions data automatically. They match spend categories to emissions factors and calculate Scope 1, 2 and 3 outputs in seconds. That turns carbon accounting from a once-a-year headache into a real-time management tool. 3. Transparent supply chains Unilever has tested AI platforms that combine satellite imagery with supply data to flag illegal deforestation in palm oil regions. If a patch of rainforest is cleared where it shouldn’t be, AI catches it fast and alerts their team. No need to wait for an audit or third-party tipoff. 4. Faster climate simulations Traditional climate models take weeks or months to run. New AI-driven models can simulate complex climate scenarios up to 25 times faster. That unlocks planning tools for city councils, small businesses and insurers who can’t wait months to model flood risks or heat exposure. Yes, AI needs energy to run. But if it helps avoid 10 times more emissions than it creates, the trade-off makes sense. So the question isn’t whether AI belongs in sustainability. It’s whether we’re serious about using every tool we have to solve the problems in front of us. At Leafr, we’ve seen consultants use AI to cut time and cost on energy audits, validate supplier claims, and surface risks early. When paired with the right human expertise, AI becomes a multiplier. Because the planet doesn’t care if a human or a machine found the emissions. It just cares that they’re found and cut. Follow Gus Bartholomew (Leafr 🌿)for more and repost if you found useful. Use Leafr to find the sustainability specialists you need to support your AI efforts

The electricity grid, the backbone of our modern world, is facing a critical challenge: integrating the rapidly growing wave of renewable energy sources. These clean energy sources are essential for combating climate change, but their intermittent nature creates hurdles for traditional grid management. This is where AI steps in as a game-changer. There is a long line of renewable energy projects waiting to connect to the grid. Delays here can significantly slow down the clean energy transition. The backlog is caused by several factors, including: ~ Manual processing of paperwork is slow and prone to errors. ~ Assessing the impact of new projects on the grid is a time-consuming process. ~ Accommodating new sources often requires expensive grid upgrades, discouraging some developers. ~ AI offers a powerful solution to these gridlock issues: ~ AI-powered tools can automate document review, flagging missing information and speeding up approvals. ~ AI algorithms can analyze vast datasets to predict potential grid impacts and suggest optimal connection points for new projects. By automating tasks and accelerating reviews, AI can significantly reduce wait times in the queue. This translates to faster integration of renewable energy and a cleaner mix of generation. AI's impact goes beyond the interconnection queue. Here's how it can transform grid management: ~ AI can help manage energy storage systems, ensuring efficient use of renewable energy even during low production periods. ~ It can analyze sensor data to predict equipment failures, preventing costly downtime and extending the lifespan of grid infrastructure. ~ In addition, AI can predict energy demand with greater accuracy, allowing for better planning and resource allocation. While AI offers immense potential, it's crucial to acknowledge the challenges. AI systems can be vulnerable to cyberattacks, requiring robust security measures. Furthermore, AI models are only as good as the data they are trained on. Biased data can lead to unfair or inaccurate outcomes. The integration of AI into the grid is no longer a question of "if" but "when." By streamlining processes, optimizing operations, and enhancing grid resilience, AI is poised to be a key driver in the transition to a sustainable energy future. With careful planning and responsible development, AI can ensure a brighter future for our power grid.

We’ve called efficiency the unsung hero of the energy transition in the past. While the energy transition will happen first through the transition of energy usages, like the shift with transport, from internal combustion engines to electric vehicles, or from fuel or gas boilers to heat pumps, we cannot ignore the utmost priority of the energy transition: efficiency. Efficiency is the greatest path to reduce our energy use, our impact on the world’s climate through CO2 emission reduction, and very importantly, the best way to make solid and practical savings. In its most historical form, energy efficiency is about better insulation, to reduce heating (or cooling) loss in buildings like family homes, warehouses, office high rises, and shopping malls. This is useful, but expensive and tedious to realize on existing installations. Digitizing home, buildings, industries and infrastructure brings similar benefits at a much lower cost and a much higher economic return. The combination of IoT, big data, software and AI can significantly reduce energy use and waste by detecting leaky valves, or automatically adjusting heating, lighting, processes and other systems to the number of people present at any given time, using real-time data analysis. It also allows owners to measure precisely progress, report automatically on their energy and sustainability parameters, and benefit from new services through smart grid interaction. And this is just the energy benefit. Automation and digital tools also optimize the processes, safety, reliability, and uptime leading to greater productivity and performance.

The next evolution of sustainable AI isn’t just about using more efficient hardware—it’s about Autonomous AI Agents that code with sustainability in mind. These agents are designed to operate independently, learning and adapting as they go, and have the potential to transform software development by writing energy-efficient code. They don't just optimize for speed; they prioritize minimal resource consumption. Why This Matters for Sustainability Modern AI models consume massive amounts of power, yet software development still prioritizes performance over energy efficiency. Agentic AI could change that paradigm by: ✅ Reducing Computational Waste: AI agents could select or generate the most efficient algorithms based on real-time constraints instead of defaulting to resource-heavy models. For example, they could optimize database queries to reduce data retrieval and processing or dynamically adjust resource allocation based on demand. ✅ Automating Green Software Principles: AI-driven frugal coding practices could optimize data structures, reduce redundant calculations, and minimize memory overhead. This could involve choosing the most energy-efficient programming language or framework for a specific task. ✅ Measuring & Optimizing in Real Time: The reward function would be clear: lower energy consumption, less latency, and reduced emissions—all while maintaining accuracy. ✅ Parallel & Distributed Optimization: AI agents could continuously refine codebases across thousands of cloud instances, improving sustainability at scale. AI-Driven Innovation Archive for Green Coding One of the most exciting ideas in autonomous coding is the "Green Code Archive"—an AI-generated repository of energy-efficient code snippets that could continuously improve over time. Imagine: 🔹 Reusing optimized code instead of reinventing energy-intensive solutions. 🔹 Carbon-aware coding suggestions for green data centers & renewable energy scheduling. 🔹 AI-driven legacy refactoring, automating migration to sustainable architectures. Measuring AI’s carbon footprint after the fact isn’t enough—the goal should be AI that reduces energy use at the source. The future of sustainable tech isn’t just about efficient hardware—it’s about intelligent, autonomous software that optimizes itself for minimal environmental impact. While this technology is still emerging, challenges remain in areas like training complexity and robust validation. However, the potential benefits for a greener future are undeniable. Learn more about leading with Agentic AI and its transformative potential in my book, "Empowering Leaders with Cognitive Frameworks for Agentic AI: From Strategy to Purposeful Implementation" (link in the comments section). #agenticai #greenai #sustainability

AI-enhanced power grid optimization can reduce emissions that are the equivalent of removing 6.5 million (U.S.) gas-powered mid-size passenger vehicles from the road for a year. “AI” is a much broader term than what most people think of—it’s not all LLMs! When it comes to reducing energy waste and operational power grid emissions, AI can help by dispatching generation assets more optimally, reducing losses, congestion, and cost. In our paper, which will be presented at the NeurIPS 2025 Workshop "Tackling Climate Change with Machine Learning," we analyze the operational emissions associated with training CANOS, Google DeepMind’s graph neural network for solving AC Optimal Power Flow (OPF) on a 10,000-bus power system. We then estimate how emissions and energy use would change if these dispatch solutions were used to determine generator (power plant) dispatch decisions, instead of the status-quo linear approximations used in many power markets to set generator output. Especially compared to training something as complex as an LLM, training these GNNs—which have a focused task (learning OPF solutions)—“pays back” all energy and emissions costs associated with the model's training within a single hour. At a country-wide scale, operating the grid more efficiently using these models is approximately equivalent to removing 6.5 million (U.S.) gas-powered mid-size passenger vehicles from the road for a year. Of course, a full analysis would require a lifecycle carbon assessment of training these GNNs. And we'd have to run the actual power grid models themselves across ISOs, not just a 10,000 bus synthetic grid. Additionally, we'd need to model other grid components and concepts like ancillary services, self-schedulers, and more. But even if we’re off by, say, a HUNDRED times, the conclusion is still clear: using a GNN approximation for dispatch can reduce energy use and emissions relative to DC OPF-based approximations. (Even if we're off by the training emissions by a *thousand* times, this holds true.) If you’re at NeurIPS in San Diego this year, please come chat with me at the session if you’re interested in this work! Read more here: https://lnkd.in/g9aqhXpy And stop saying "AI" when you actually mean LLMs. :)

🔍 What’s the net climate impact of artificial intelligence? I just spent the last three hours reading a compelling new thesis from Jennifer Turliuk, introducing the Net Climate Impact Score (NCIS)—a novel framework for assessing whether AI reduces or increases greenhouse gas emissions when viewed holistically. The research explores the dual role of AI as both a climate challenge and a potential climate solution. I am a big fan of Jennifer - for those unaware she is practice leader of climate and energy AI at the Martin Trust Center for MIT Entrepreneurship, developing new AI tools in collaboration with the MIT Climate Policy Center to support Climate and Energy Ventures and she recently led a panel at Davos 2025 on how AI can accelerate the energy transition. Her thesis researches how AI is rapidly expanding across infrastructure, industry, and digital services. But its environmental footprint—through energy-intensive data centres, chip manufacturing, water use, and end-user applications—is growing faster than most realise. Some projections suggest that AI and its supporting infrastructure could consume up to 21% of U.S. electricity by 2030, placing AI in direct tension with corporate net-zero targets and increasing humanity's need for energy at a time when we should be reducing. The paper is built around a system dynamics model. It highlights the potential rebound effect of AI - that as AI becomes more energy efficient (e.g., via better chips or software), it becomes cheaper and more accessible. This often leads to increased overall use, offsetting the efficiency gains. These effects can be : - Direct rebound effects: Lower costs per AI task drive higher usage across sectors. - Indirect rebound effects: Efficiency frees up resources that are reinvested in other high-emission activities, compounding total climate impact. The NCIS framework balances: 🔴 AI’s climate harms (emissions across AI infrastructure, plus enabling fossil fuel exploration), and 🟢 AI-enabled emissions reductions (e.g., optimised grid operations, predictive maintenance, smart EV charging). 🧠 Key insights: = AI’s potential to cut emissions is 1.5–4% by 2030 (PwC, IEA, BCG estimates). = However, the actual emissions from AI are growing and may outweigh benefits without targeted deployment and regulation. = The time value of carbon means that emissions today are more damaging than potential savings tomorrow. 📌 Bottom line: AI 'can' support decarbonisation only if strategically aligned. Industry leaders need to: 1. Prioritise AI for climate-beneficial use cases (e.g., energy, transport, buildings). 2. Monitor NCIS-style metrics when evaluating AI investments. 3. Advocate for policies that internalise carbon costs and drive clean computing. We’re at a fork in the road. Will AI accelerate the climate transition—or become a high-emitting enabler of business-as-usual? You can get early access to the report here: https://lnkd.in/gwJQ8MjW

The #AI era and the energy transition are advancing rapidly –– but unequally. 565 million people in Sub-Saharan Africa still lack electricity, and 900 million rely on biomass. Whilst annual data center spending to fuel the AI boom could require investments up to USD 7 trillion by 2030. But improving the alignment of these 21st century trends could lead to important impact. When planned together, digital infrastructure and AI-enabled innovation can anchor new renewable generation, create reliable offtake, and catalyze local innovation ecosystems. What does AI mean for development if it does not change how people access energy, cook, or economic opportunities? During a Ministerial Address at the International Renewable Energy Agency (IRENA) Assembly, I delved into what development-centered and meaningful AI-energy alignment looks like: 1️⃣ AI is accelerating renewable deployment by turning planning into precision delivery: AI-enabled forecasting, digital twins, and grid analytics improve dispatch accuracy by 20-30% and cut curtailment by 10-12%, enabling faster integration of wind and solar at scale. 2️⃣ Digital systems are converting efficiency gains into fiscal space for access and inclusion: Energy inefficiency still drives nearly 60% of global energy losses, costing developing economies up to USD 1.3 trillion by 2030. Even a 10% efficiency gain, enabled by AI-based demand management and predictive maintenance, can unlock billions in savings. 3️⃣ Livelihood creation through productive use of energy: When energy and digital transitions converge, they create jobs, incomes, and development. And as clean energy is combined with digital tools, data, and market intelligence, it becomes a platform for economic transformation. With this emerging technology, we have the opportunity to revolutionize our energy systems, but only if AI is deliberately aligned with access, affordability, and resilience, rather than layered onto existing inequalities.

Since 2000, the number of humans worldwide without access to electricity has dropped by half -- nearly 1 billion more people have electricity today. Several factors contributed to this improvement, but a key driver is technology. Rapid technology development has transformed energy systems over the last two decades. And today, AI is poised to accelerate this trend in unprecedented ways. Recent insights from Reuters highlight several uses of AI to advance and improve our power systems. Let's take mini-grids as an example. Mini-grids (similar to micro-grids) harness local storage and generation, creating decentralized networks that empower communities, particularly rural ones. Imagine a small community in the mountains installing solar panels and batteries, forming a local energy generation and storage facility to power their needs. This shift from centralized authorities to local innovators accelerates electrification, sidestepping the prolonged process of waiting for regulatory approvals and building new power plants. A key challenge micro-grids face is managing supply & demand for users. But now, predictive AI is enabling innovators like Husk Power Systems to forecast supply and demand, delivering electricity at the cheapest point at any given time. Similarly, installing and managing generation assets can be a challenge especially in rural environments. Here, AI-based models can optimize site selection & improve predictive maintenance for wind turbines and solar panels. By making mini-grids more efficient, AI is accelerating rural electrification -- just one example of how such technologies can enable cheaper, cleaner, and more reliable power for millions around the world. 🌍 #AI #aiforgood #energysystems #sustainability #minigrids #microgrids #cleanenergy #energytransition #cleantech #climatetech

How can AI help us create more sustainable energy? Our recent research offers a compelling answer. Our team has been applying machine learning to solve some of the trickiest challenges in biofuel production. In our first paper (https://lnkd.in/gCnib5Nk), we used a Random Forest model to uncover a surprising truth: iron and phosphorus are more crucial for maximizing algal lipid yields than the long-held focus on nitrogen starvation. This data-driven approach allowed us to reduce the necessary experimental conditions by over 96%. Now, our latest paper (https://lnkd.in/gU8s736n) takes this even further. We used machine learning to understand how a virus can act as a metabolic modulator for algae, boosting lipid production by over 165%. This bypasses the need for today's high-energy, costly extraction methods. Access the full paper for free for the next 50 days here: https://lnkd.in/gKQHYUiH. This is a fantastic example of how AI can lead us to entirely new, biomimetic strategies for a greener future!

The International Energy Agency (IEA) recently published its World Energy Outlook Special Report, a comprehensive look at the links between energy and AI.    I recently shared how AI will become an important tool for building energy and climate solutions with Fortune Magazine (link in comments). And, after reading the IEA report, I’m more convinced than ever that AI has an important role to play in unlocking efficiency and operational gains for the energy sector.    The IEA report includes several examples: Optimizing the integration of variable renewable energy sources into the grid: AI can be used to improve the forecasting and integration of wind and solar photovoltaic generation, reducing curtailment and associated emissions. For example, DeepMind's wind power forecast was found to potentially increase the financial value of wind energy by as much as 20%. Reducing the global average curtailment by just one percentage point in 2035 could prevent approximately 120 metric tons of CO2 emissions.   Reducing methane emissions in the oil and gas sector: AI can boost data processing techniques to detect and quantify methane emissions from leaks, enabling continuous monitoring at a larger number of facilities and pipelines, potentially leading to significant emissions reductions. In fact, implementing continuous leak detection and repair could avoid nearly 2 metric tons of methane emissions globally.    Optimizing energy use in the transport sector: AI applications such as route optimization, predictive maintenance, and improved capacity utilization can cut energy consumption in road, air, shipping, and rail transport. Widespread adoption of existing AI applications across transport modes could save over 4.5 exajoule of energy by 2035.  For example, AI-driven flight route optimization systems have the potential to reduce fuel consumption by 5-12% per flight.    These are just three of the examples listed in the report, which I urge you to check out here: https://lnkd.in/gmvZxrJ4 While there’s no silver bullet when it comes to AI, energy and the environment, at Crusoe, we're dedicated to an energy-first approach in building AI infrastructure. This commitment fuels our optimism as we work towards a future where AI innovation and a cleaner, more efficient energy landscape go hand in hand.