Most people think of the internet as weightless. In reality, every email, video stream and machine learning query relies on physical infrastructure that consumes energy and generates greenhouse gases. The footprint of our online lives is growing, driven by factors that are technological, economic and behavioral. Understanding these drivers helps organisations and individuals focus their efforts where they will have the most impact.
What we mean by digital carbon emissions
Digital carbon emissions are the greenhouse gases released across the lifecycle of digital services. That includes the electricity used by data centres and network equipment, the energy embedded in manufacturing devices, the emissions linked to cloud services and the operational energy for software and AI models. The term covers both direct power consumption and upstream impacts from materials, production and transport.
Whats driving the rise in digital emissions
Several interlocking trends have expanded digital demand and the energy needed to serve it. First, the sheer volume of data traffic continues to grow as people stream higher-resolution video, participate in online gaming and use rich, interactive apps. Higher quality media and always-on services require more network capacity and more compute to process, store and deliver content.
Second, compute-intensive workloads have proliferated. Large-scale artificial intelligence training, real-time analytics and personalised services rely on clusters of specialised hardware running for long periods. These workloads are powerful enablers of innovation, but they also drive sustained increases in electricity consumption unless efficiency improvements outpace demand.
Third, device proliferation multiplies energy draw and embedded emissions. Households now own multiple internet-connected devices, and new categories of smart hardware increase the number of endpoints that need manufacturing, powering and eventual disposal. The cumulative environmental cost of producing and replacing billions of devices adds to digital-sector emissions even when per-device energy use falls.
Fourth, cloud adoption and decentralised compute create complex consumption patterns. Cloud platforms can be efficient at scale, but shifting workloads between providers, geographic regions and delivery models can move emissions rather than eliminate them. Without active optimisation and clean energy procurement, migrating workloads to the cloud can increase total energy use.
Finally, user behaviour and expectations play a role. Instantaneous responses, personalised feeds and endless autoplay features increase time spent online and the number of server requests per session. Design choices that prioritise experience over efficiency amplify the energy footprint of routine activities.
Why this trend matters beyond kilowatt-hours
The growth of online emissions matters for several reasons. From a climate perspective, digital services are part of the global emissions picture and need to be included in credible decarbonisation strategies. For businesses, rising digital emissions can affect regulatory compliance, investor scrutiny and reputational risk as stakeholders increasingly expect transparent climate performance across all operations, including digital.
There is also a fairness and access dimension. Regions with carbon-intensive electricity grids will see greater greenhouse gas consequences from equivalent digital use than places powered by low-carbon electricity. This creates inequities in the climate cost of identical services and complicates corporate claims about ‘green’ digital products unless regional energy sourcing is accounted for.
Finally, digital emissions are intertwined with material flows and waste. Device manufacturing consumes raw materials and energy, and frequent replacement cycles increase electronic waste. Addressing digital emissions therefore touches manufacturing, supply chain, and end-of-life management, not just server efficiency.
Where organisations can make the biggest difference
Companies that build, run or consume digital services have several levers that can reduce emissions in meaningful ways. The first lever is energy sourcing: choosing suppliers or data centres powered by low-carbon electricity directly reduces the operational emissions tied to compute. Purchasing long-term renewable energy or matching consumption with credible renewable attributes moves the needle more than one-off offsets.
Second, product design and software efficiency are powerful and often underused tools. Optimising code paths, reducing redundant network calls, deferring heavy processing to less carbon-intensive times or locations, and trimming media bitrates where appropriate all cut energy demand without changing user value. Small, habitual savings at scale translate to significant emissions reductions.
Third, infrastructure choices matter. Selecting more efficient hardware, consolidating workloads to reduce idle servers, using specialised accelerators only where they offer clear efficiency benefits, and applying better cooling methods in data centres reduce energy intensity per unit of work. Capacity planning that avoids unnecessary overprovisioning also helps.
Fourth, measurement and reporting create accountability. Clear metrics that connect digital activity to emissions allow teams to prioritise interventions and demonstrate progress to stakeholders. Transparent reporting also helps avoid accusations of greenwashing by making assumptions and boundaries explicit.
Practical steps individuals can take
Individual actions are meaningful when aggregated across millions of users. Simple behaviours such as lowering streaming resolution when high definition is unnecessary, closing unused browser tabs and apps, and extending the lifespan of devices reduce both operational and embodied emissions. Choosing services and providers that publish their energy sourcing and efficiency practices helps shift market demand toward lower-impact options.
More engaged users can influence product choices by providing feedback to platforms about features that feel wasteful and by favouring apps that offer energy-saving modes. For many people, the most impactful step is aligning consumption habits with values: reducing needless uploads, limiting rarely used cloud backups, and resisting frequent device upgrades all curb the demand that drives emissions upstream and downstream.
Policy and market mechanisms that accelerate change
Public policy and market signals can accelerate digital decarbonisation. Standards for reporting digital emissions, procurement rules that favour lower-carbon providers, and incentives for data centre efficiency improvements help create a level playing field. Carbon pricing policies that include digital-sector emissions would internalise climate costs and make efficiency investments more attractive.
Procurement choices by large organisations also matter. When major companies require verifiable renewable energy for cloud footprints or include digital lifecycle emissions in procurement criteria, providers respond by improving transparency and investing in cleaner operations. Investor scrutiny and consumer pressure together can be a strong force for change.
Balancing innovation and responsibility
Digital innovation delivers economic and social benefits, but it does not have to come at the cost of rising emissions. The trick is to pair technological progress with explicit efficiency and sourcing strategies. Engineers, product managers and procurement teams can embed emissions considerations into decision-making so that new features and capabilities are assessed not only for functionality and cost, but also for climate impact.
Ultimately, the goal is not to hinder innovation but to shape it. With the right incentives and practices, digital services can continue to expand access and productivity while growing more energy-efficient and less carbon-intensive.
Understanding why digital emissions are increasing is the first step. The next step is practical: focus on where demand can be reduced or met with cleaner energy, design products that do more with less, and adopt measurement practices that make progress visible and verifiable. That combination keeps the benefits of the digital economy while aligning it with broader climate goals.