Brace Yourself for 2025: Key Trends in Climate and Sustainability

2025 is here. Corporate Sustainability regulations are kicking in (or still kicking it in the courts). AI and energy demand only goes up. There are massive amounts of renewable energy lined up to get connected to the US grid. And even as the clock runs out on the Biden Administration, the US federal government is racing to lock in more spending, rules, and protections in an effort to preserve ground gained via the Inflation Reduction Act and other wins.

Meanwhile, the global data center boom shows no signs of slowing down, driving innovation but also sparking hard questions about energy and water usage. And yes, many organizations will miss their 2025 guidepost emissions reduction targets, but that’s not the end of the story.

Let’s explore what’s ahead for 2025 and how we can navigate the changes in the coming year.

Prediction #1: CSRD reporting starts in 2025, but making sense of the reports will be a serious challenge.

📜 There’s no standard format for documents, no central repository, and the technical tagging standards for data are a little esoteric. Reports will be hard to find when they’re published, hard to parse consistently, and hard to interpret, given the inconsistencies. 

📊 And maybe that’s fine, for now? The EU’s CSRD requirements force some of the world’s largest companies, including some in the US, to measure and report GHG emissions and other key sustainability metrics. 

📚 If that’s what forces a strong first step for some organizations, we should be at least pleased, if not necessarily thrilled. The first audience for these reports should be the companies doing the reporting themselves, which will now have actionable data to plan carbon emissions reductions projects. 

📉 The second audience might be stakeholders, stockholders, and regulators, who can look at any one company in isolation and understand how their emissions change year-over-year. 

🤔 That said, making enough sense of the numbers in these reports to compare one company to another might be a tough ask. Different standards, jurisdictions, and definitions of what’s material will make comparisons challenging. 

👋 Expect some well-intentioned hand-waving and caveats from everyone who tries to glean and share useful insights from these first CSRD reports.

❓ Where does that leave us, and what would be a better solution? It will be at least a couple more years before the EU has a useful central repository for these reports, which may or may not parse and aggregate the data in any useful way. 

💡 There’s probably an opportunity for someone willing and able to gather the reports as they’re released, summarize them, and extract the key data to turn into comparison tables and charts over time.

Prediction #2: AI will not go away; demand for electricity to power AI will grow.

🚫 Don’t make the mistake of classifying Generative AI as a fad. (Yes, I am talking to you.)

🏒 The Generative AI tools that hockey-sticked to prominence in 2023 and 2024 are *not* going anywhere. As you’ve hopefully heard by now, AI is not coming for your job – people who know how to use AI are coming for your job. And there are many more serious uses for GenAI than asking ChatGPT to write a sonnet about pizza, or a limerick about pubs in London with funny names. (Not that I... have done either of those... recently...)

⚡️ And yes, Generative AI uses more electricity and energy than a Google search, but so does any new use of computing technology that is still evolving and needs optimization.

💰 We have a lot of work to do; soon, the hyperscalers and foundational model builders will have to spend time optimizing AI for energy efficiency as a cost control measure, with hundreds of billions of dollars in capital being poured into AI without any meaningful trend toward profitability (yet).

💡 Optimizing AI for energy efficiency could be a differentiator for some providers of software platforms and services in 2025.

🔋 Wrapping open-source AI models in sustainable software and data centers could also be a huge opportunity for someone in 2025.

Prediction #3: The data center boom will continue unabated.

🚜 Thanks largely to the aforementioned growing demand for AI, and partly to the ongoing boom in software, information processing, and automation, new data centers will continue to come online, break ground, and get planned with no end in sight.

😎 The hyperscalers will lead from the front, and some innovations (chip-level cooling, anyone?) will trickle down to other vendors.

⚙️ We're going to see more diversification and specialization in data center construction. The requirements and usage patterns of AI are different from those of blockchain, which are different from those of high-traffic website publishers.

🟢 Some of the more ambitious data center builds in the coming years will place computing power adjacent to renewable energy resources.

🍃 The idea of siting a data center inside a wind turbine is a few years old now, but the premise of a (nearly?) “off the grid” data center running on renewable power is still a strong one.

🛢️ Hydrotreated Vegetable Oil (HVO – think biodiesel, but a little cleaner) can power backup diesel generators, helping more data centers go fully renewable with less reliance on natural gas when the sun isn’t out and the wind doesn’t blow.

Prediction #4: Everyone is going to miss their 2025 sustainability targets. Don’t panic. Yet.

😬 Record scratch, freeze frame… Yep, that’s thousands of the world’s biggest companies reporting their carbon emissions numbers in 2025 thanks to the EU CSRD, along with many that were already reporting. They’re mostly going to fall short of their goals for 2025, which was supposed to be an important (but arbitrary) signpost on the way to 2030, the (arbitrary) deadline to prevent a slip past a 1.5°C increase in average global temperatures over the pre-industrial numbers.

😅 Welp, there’s good news and bad news. The good news is that *everyone* is behind on their goals, it’s not just you. The bad news, is we’re going to careen through that 1.5°C increase in global temperatures long before 2030. (As of this writing, it’s now official that average temperatures crossed that line in 2024, though it will take a little longer before the thirty-year average does so. But not much longer, unless…)

😱 Don’t panic. Yet. If you’re not measuring your emissions, start measuring. If you’re measuring but not reducing emissions yet, start reducing your emissions. Missing 2025 targets shouldn’t be seen as a failure of sustainability efforts – it should be seen as an opportunity to differentiate your company by making a measurable difference, and soon.

Prediction #5: The US federal government will backslide on climate efforts, but will stop short of undoing the progress of the Inflation Reduction Act.

Donald Trump will be the next President of the United States in just a few days. We can’t undo that. But, as dire as things seemed for sustainability the morning after the 2024 election, the Biden administration’s work to pass and implement as much as possible of the Inflation Reduction Act (IRA) before Jan. 20, 2025 will pay off. 

It’s the closest thing to a “Green New Deal” we’ve ever seen, bringing tens of billions of dollars of funding to renewable energy efforts. Much of this funding has already been allocated, reaching a critical point of no return. The Trump administration can’t undo that, and they might not want to; many states that leaned heavily toward Trump in the 2024 election are also key beneficiaries of these investments, making widespread rollbacks unlikely.

Can we expect any CSRD-like regulations to take effect before 2029? No. Can we expect individual states to pass their own regulations? Yes. Can we expect the courts to uphold them? Maybe! 

But whether or not regulations requiring corporate sustainability reporting are in play, the funding won by the Biden administration will continue to have a positive effect on climate change for years to come.

Here's how we'll approach 2025...

This year will test our resolve, creativity, and our ability to adapt in the face of growing climate challenges. Expect more disasters. Expect less help from the US federal government. Expect bureaucratically complex carbon emissions standards and reporting.

But positive outcomes are possible. We can use AI to improve energy efficiency, we can experiment with microgrids and self-contained electricity solutions for data centers, and we can turn missed goals into opportunities to reset corporate sustainability efforts to better match the problem at hand. There’s no shortage of work to be done.

Want to get started? We can help.