Coding Climate Resilience - Can Technology Actually Support a Just and Sustainable Future?
The 1st in our Sustainability Explainer Series by Peter Simpson
At Ozyntel, we hold complexity like a positive obsession - as Octavia Butler put it; researching how to work ethically with technology being one of many tendrils of tension we work to reflexively process into navigational intelligence for people and planet.
We are acutely aware of and concerned about the massive environmental impact of AI data centres, whilst observing reputable research centres, such as Stockholm Reslience Centre, working with Google Deepmind to produce their recent metareview of 8500 articles in this area, ‘AI for a Planet Under Pressure’.
We want to confront and navigate this paradox as we move past the edges of our own certainty and know-how, whilst translating what we find into something intelligible and explainable for people outside of academic, tech, or policy bubbles.
This article is the first in our ‘Sustainability Explainer’ series that delves into the jargon and opacity of sustainability, COPs, climate policy, and tech’s influence on all the above, initiated by our Research Intelligence & Project Analyst, Peter Simpson. We hope you find it helps create some clarity and insight in a turbulent world. Resources for further reading, as ever, can be found beneath.
As the climate talks in Belém reached their final hours, negotiators were arguing over fossil fuel phase‑out language, finance commitments, and what “just transition” really means in practice. Beneath the headline debates sits a quieter question that will shape whether any of this holds: what kind of digital infrastructure are we building to deliver on these promises?
This question must be asked, and directly dealt with; technology has always been both a culprit and a catalyst in the sustainability story. On one hand, it drives consumption, fuels e‑waste, and powers energy‑hungry data centres. On the other, it helps us monitor forests in real time, model coastal flooding decades ahead, and coordinate decarbonisation across entire supply chains. The question is no longer whether technology can advance sustainability. It is how we design, deploy, and govern it so that it supports justice, resilience, and systemic change rather than accelerating extraction.
This piece is part of an ongoing attempt to translate the sometimes dense language and complexity of sustainability, climate policy and the huge influence of Big Tech into everyday questions: who wins, who loses, and who decides how we programme the systems that will shape our future.
The Paradox of Progress
Every new wave of digital innovation arrives with a familiar paradox: more capability, more risk.
Artificial intelligence (AI) helps optimise energy systems and forecast extreme weather, yet training and running large models can consume as much electricity as a small town.
Cloud computing makes collaboration easier across borders, but hyperscale data centres concentrate energy demand and water use in specific regions, often where communities already face heat and water stress.
High‑resolution climate models and satellite imagery improve our understanding of risk, but they also depend on infrastructures that lock in new dependencies on private platforms and proprietary data.
For organisations that tell themselves they are “going digital” to be more sustainable, this raises uncomfortable questions:
Are our digital operations actually energy‑efficient, or are we simply shifting emissions from one part of the balance sheet to another?
Are we collecting and analysing data in ways that measurably improve sustainability outcomes, or just generating dashboards and press releases?
Are we building resilience into our systems, or only optimising for short‑term efficiency and convenience?
These are not just abstract questions confine to Belém, and other fora, such as the World Economic Forum in Davos. They show up when negotiators discuss early‑warning systems for extreme weather, digital monitoring of deforestation, or the intelligence needed to esatablish and implement loss and damage mechanisms1. Going beyond mere slogans, they sit in black and white in the Paris Agreement, the Warsaw International Mechanism and the latest COP decisions. The tools we choose, and how we govern them, will quietly decide who gets protected, who remains exposed, and who actually gets compensated for the baked-in damage.
The Tools We Reach For
Let’s briefly translate some of the technologies that show up in climate and sustainability conversations, without the jargon.
Artificial Intelligence
AI is becoming a core part of sustainability analytics - we can give concrete use cases for where its presence can bring positive impacts depending on how, and with what intent and awareness, it is used:
Detecting deforestation and illegal mining from satellite imagery.
Optimising energy grids as renewables fluctuate through the day.
Forecasting floods, droughts, and heatwaves so that communities can prepare.
When AI is designed and governed well, it can support climate resilience: earlier warnings, smarter infrastructure planning, and better use of scarce resources. When it is not, it can amplify bias, hide uncertainty, and concentrate power in a handful of companies whose models become de facto infrastructure for public decision‑making.
Internet of Things (IoT)
IoT simply means networks of connected sensors and devices:
Monitoring air quality, water levels, and soil moisture in real time.
Managing traffic flows and public transport to cut congestion‑related emissions.
Tracking building performance so energy savings move from policy documents into practice.
In principle, this data helps cities and regions move from reactive crisis response to proactive risk management. In practice, without clear governance and capacity on the ground, sensor networks can become expensive pilot projects that never translate into better lives.
Blockchain and Traceability
Distributed ledgers are often mentioned in the context of:
Verifying that raw materials come from certified, sustainable sources.
Tracking products through complex global supply chains.
Providing tamper‑resistant records of climate finance flows.
The promise here is transparency. The risk is that the energy use of some blockchain systems, and the complexity of their implementation, can outweigh the benefits if we are not careful about design choices and standards.
Big Data and Cloud Platforms
Large‑scale data platforms underpin:
Corporate sustainability reporting and scenario analysis.
National greenhouse gas inventories and progress tracking.
Global “stocktakes” of where we stand against long‑term climate goals.
These platforms can make sustainability targets real by tying them to measurable indicators and timelines. But they also raise questions: whose data is included, whose reality is visible, and who has the skills and access to interpret the numbers?
Policy programmes like the European Green Deal and Destination Earth imagine exactly this kind of sensor‑rich, model‑driven infrastructure2. In sum, who has access to digital infrastructure, skills, and value capture, as both UNCTAD and UNDP show, hews closely to existing economic and climate vulnerability fault lines3
Energy, Ethics, Equity: Three Tests for Assessing “Sustainable” Tech Claims
The ILO’s just transition guidelines remind us that decarbonisation pathways are ultimately about jobs, rights and social protection as much as technology4. If we want technology to support a just transition rather than derail it, we can use three simple assessments, or mental tests - Energy, Ethics, Equity - for any digital solution that appears in climate plans, corporate strategies, or COP declarations that we read about.
1. Energy: What does this actually cost the planet?
The global information and communications technology (ICT) sector is estimated to be responsible for around 3–4% of global carbon emissions, and that share is rising (Freitag et al, 2021). As several analyses of ICT energy use suggest, our digital systems, such as data centres, already account for a non‑trivial slice of global emissions5.
Add to this:
Growing electricity demand from AI training and inference.
Cooling requirements for data centres in increasingly hot regions.
Water footprints in locations already facing scarcity.
A “sustainable” digital system should:
Be designed with strict energy budgets and efficiency targets, not treated as an infinite, invisible resource.
Prioritise green cloud computing: data centres powered by renewables, sited and operated with local ecosystems and communities in mind.
Make its energy and resource use transparent enough that regulators, users, and affected communities can scrutinise it.
If climate resilience is the goal, then infrastructure that becomes unusable during heatwaves or grid stress is not “smart”, no matter how advanced the AI. Recent work by the ITU, SRC, WMO and OECD documents dozens of AI‑for‑climate use‑cases, from flood forecasting to grid optimisation6.
2. Ethics: Who controls the models that guide our decisions?
Ethics in this context is not a checklist. It is about power and accountability in systems that increasingly shape climate policy and investment.
Key questions:
Whose models and datasets are used to project risks, design adaptation strategies, and allocate climate finance?
How transparent are the assumptions baked into these systems—and who can challenge them?
What happens when proprietary platforms become the default infrastructure for public decision‑making?
Ethical digital systems for climate action should:
Be auditable: independent experts can scrutinise methods, biases, and limitations.
Respect data rights and privacy, especially for communities under surveillance in the name of “monitoring”.
Avoid turning climate‑vulnerable regions into sources of raw data with little say over how that data is used: a pattern sometimes described as data colonialism.
Answering questions of whose knowledge and datasets dominate - ‘nothing about us, without us’! - are thus crucial. As Nick Couldry and Ulises Mejias argue, much of today’s data economy reproduces older colonial patterns of extraction and control7. Without acknowledging and acting on this, we risk a future where “AI says so” quietly overrides local knowledge, democratic debate, and precaution.
3. Equity: Who benefits, who bears the risk?
Equity connects directly to loss and damage, just transition, and climate finance - recurring points of friction at recent COPs, and frequently highlighted by the latest climate science from the IPCC.
Consider:
Many climate‑vulnerable countries still lack robust digital infrastructure, reliable connectivity, or the human capacity to use complex tools.
Yet these same countries are often expected to provide high‑quality data, adopt digital monitoring systems, and participate in global climate reporting frameworks.
Meanwhile, most of the economic upside from digital innovation in climate tech flows to firms and investors in wealthier economies.
A fair digital transition would:
Treat digital access and capacity as part of climate resilience, not a luxury add‑on.
Link climate finance to concrete investments in open, locally governed data systems and tools.
Ensure that communities closest to the frontlines of climate impacts have a meaningful role in designing and governing the digital infrastructures that affect them.
If equity is not built in, digital tools risk hard‑coding existing inequalities into the systems that decide who receives support after a disaster, where adaptation funds flow, and whose losses are recognised. IPCC AR6 is clear that resilience depends as much on social and institutional capacity as on the sophistication of our models8 (See also SRC, AI for a Planet Under Pressure, 2025).
From Harm Reduction to Regeneration
Sustainable technology is often framed as an exercise in harm reduction: lower emissions, less waste, more efficiency. Necessary, but insufficient.
The deeper challenge is to design digital systems that actively support regenerative futures; systems that help restore balance instead of simply slowing damage.
That could look like:
Circular electronics models where devices are designed to be repairable, modular, and recoverable, so that critical minerals are used more than once instead of becoming toxic waste.
Digital twins for cities, river basins, and industrial systems that allow us to test scenarios - such as heatwaves, floods, or supply chain shocks - before we build, so that resilience is designed in from the start. The EU’s ‘Destination Earth’ initiative is already building digital twins of climate and infrastructure systems to test such scenarios before they unfold in the real world [1].
Knowledge platforms that connect peer‑reviewed science, policy commitments, and lived experience in ways that are accessible beyond expert communities, helping decision‑makers and citizens alike navigate complexity.
Here, the promise of AI, IoT, and data platforms is not just smarter optimisation. It is learning systems that help us notice early warnings, avoid lock‑ins, and adapt together, building capacity as conditions change.
Belém as a Test of How We Programme the Future
As the talks in Belém close, the familiar patterns are playing out:
Disputes over how strongly to commit to phasing out fossil fuels.
Battles over the scope and structure of loss and damage mechanisms.
Tension between near‑term national interests and long‑term planetary stability.
Threaded through these is a quieter set of decisions about climate just digital infrastructure:
Will early‑warning systems reach those most at risk, or remain pilot projects in a handful of countries? As UNDRR’s work on the human cost of disasters makes plain, early‑warning coverage is still heavily skewed towards those already better protected9.
Will monitoring and reporting systems be open and interoperable, or tied to proprietary platforms?
Will climate finance support digital capacity in the places that need it most, or primarily subsidise new markets for private technology providers?
These are, ultimately, questions about how we are programming our shared future, and what values are compiled into the code, contracts, and institutions that will outlast any single summit.
A Call to Programme for Resilience and Justice
So perhaps we can retire the question, “Can technology save the planet?” It is too vague, and it lets us outsource responsibility to tools.
The more honest questions are:
Which technologies are we choosing to scale, and on whose terms?
Whose resilience are we prioritising when we design digital systems for climate action?
How do we ensure that the energy, ethics, and equity tests are applied not just in pilot projects, but in the infrastructure that COP decisions quietly lock in for decades?
If we code with conscience, innovate with empathy, and prioritise long‑term resilience over short‑term efficiency, technology can become not the enemy of sustainability, but one of its most powerful allies.
The outcome in Belém will be read in years to come not only in the headline pledges, but in the infrastructures and institutions that emerge from them. The real work starts when we translate those words into the newly constructed systems - digital and otherwise - that shape and support everyday life.
The question is no longer whether technology can support a just and sustainable future. It is whether we will choose to programme it that way. We all have agency - the extent to which we can actualise it depends on our power, privilege and platform, as well as the knowledge and technology we may have at our fingertips.
UNFCCC (2015), Paris Agreement (Articles 7 and 8);
UNFCCC (2013–2023), decisions under the Warsaw International Mechanism for Loss and Damage and the Santiago Network;
UNFCCC (2023), Decisions adopted at COP28 / CMA5 (Global Stocktake, Just Transition Work Programme, and Loss and Damage Fund) —> Decisions portal: https://unfccc.int/documents (filter by “COP28” and “CMA5”)
European Commission (2020–2023), A European Green Deal and Shaping Europe’s Digital Future communications;
European Commission / ESA (2022), Destination Earth (DestinE) initiative documents; related digital‑twin pilots for cities and industrial systems in EU research programmes.
UNCTAD (2021, 2022), Digital Economy Report (chapters on data, development and digital divides);
UNDP (2021), The State of Climate Ambition and related briefs on digitalisation, inequality and resilience.
ILO (2016), Guidelines for a just transition towards environmentally sustainable economies and societies for all; see also subsequent ILO just‑transition briefs. https://www.ilo.org/topics-and-sectors/just-transition-towards-environmentally-sustainable-economies-and-societies
Anders Andrae & Tomas Edler (2015), On Global Electricity Usage of Communication Technology: Trends to 2030, Challenges (MDPI).
See also Data Centres and Data Transmission Networks – Analysis, International Energy Agency - https://www.iea.org/energy-system/buildings/data-centres-and-data-transmission-networks.
And especially: Freitag, P., Berners‑Lee, M., Widdicks, K. et al. (2021), The real climate and transformative impact of ICT: A critique of estimates, trends and regulations, Patterns 2(9).
ITU & WMO (2023), AI for Climate Action (AI for Good initiative);
Stockholm Resilience Centre (SRC), Galaz, V., Schewenius, M., Donges, J. F., Fetzer, I., Zhivkoplias, E., Barfuss, W., ... & Vadrot, A. (2025). AI for a Planet Under Pressure.
WMO (2023), Early Warnings for All – Executive Action Plan 2023–2027;
OECD (2022), AI in the Context of Climate Change.
Couldry, N. & Mejias, U. A. (2019), The Costs of Connection: How Data Is Colonizing Human Life and Appropriating It for Capitalism, Stanford University Press.
IPCC (2022), Climate Change 2022: Impacts, Adaptation and Vulnerability (AR6 WGII), especially Summary for Policymakers and Chapters 8, 16, 17.
UNDRR & CRED (2020), The Human Cost of Disasters: An Overview of the Last 20 Years; UNDRR briefs on risk‑informed early‑warning and resilience.
 | A guest post by
|