The Environmental Cost of Computing
Digital technology has significant environmental impacts, from the raw materials used to make devices, to the energy consumed running them, to the problem of disposing of them at end-of-life. While computing also offers environmental solutions, the overall footprint is substantial and growing.
50M+
Tonnes of e-waste generated globally each year
~2%
Global electricity used by data centres
3-4%
Global CO₂ emissions from ICT sector
10yr
Average lifespan of a well-maintained laptop vs 3 years in practice
E-Waste (Electronic Waste)
☠️ The e-waste problem
- E-waste = discarded electronic devices: old phones, computers, TVs, tablets
- Contains toxic materials including lead, mercury, cadmium, and brominated flame retardants
- When sent to landfill, these toxins leach into soil and groundwater — poisoning ecosystems
- Much e-waste is illegally exported to developing countries where workers strip components under dangerous conditions without protective equipment
- Planned obsolescence: manufacturers deliberately limit device lifespans (e.g. software updates that slow old hardware, non-replaceable batteries) to drive repeat purchases
- Short product cycles create huge volumes of waste — a new phone model every year
♻️ Reducing e-waste
- Right to Repair movement: campaign for laws requiring manufacturers to provide spare parts and repair manuals — the UK Right to Repair regulations require it for some appliance categories
- Certified e-waste recyclers: reputable organisations safely extract materials (gold, copper, rare earths) rather than sending to landfill
- Buying second-hand devices and refurbished hardware
- Designing for longevity: modular devices (like Fairphone) allow component replacement
Energy Consumption of Data Centres
Data centres house the servers that run cloud services, streaming platforms, social media, AI models, and the internet. They consume enormous amounts of energy — both to power servers and to cool them (servers generate significant heat).
| Source of energy demand | Impact | Mitigation |
|---|
| Server computing | Thousands of servers running 24/7 at high power draw | Energy-efficient processors; switching off idle servers |
| Cooling systems | Often equals or exceeds computing power use (PUE) | Natural cooling; locating in cold climates (Iceland, Norway); liquid cooling |
| AI training | Training a large AI model can use as much energy as flying from London to New York thousands of times | More efficient architectures; use of specialised AI chips (TPUs) |
| Cryptocurrency mining | Bitcoin network uses more electricity than many countries | Proof-of-stake alternatives use far less energy than proof-of-work |
Many major cloud providers (Google, Microsoft, Amazon) have pledged to achieve carbon neutrality by purchasing renewable energy certificates and investing in solar/wind power. However, the sheer growth in data centre energy demand means absolute emissions may still rise even as renewable percentages increase.
Carbon Footprint of Digital Activities
Every digital action has a small but real carbon cost — the electricity used by servers, network infrastructure, and the user's device. At scale, this adds up:
- Streaming one hour of HD video: approximately 36g CO₂ (similar to boiling a kettle twice)
- Sending one email with a large attachment: ~50g CO₂
- A Google search: ~0.2g CO₂ (low individually, but billions per day adds up)
- Training a large language AI model: estimated hundreds of tonnes of CO₂
Manufacturing Environmental Impact
⛏️ Raw material extraction
- Rare earth minerals (lithium, cobalt, tantalum, coltan) are essential for batteries and circuit boards
- Mining is environmentally destructive: habitat destruction, soil erosion, water pollution, and high energy use
- Some mines operate in conflict zones with poor labour conditions (e.g. Democratic Republic of Congo cobalt mining)
- Manufacturing a single smartphone generates approximately 70kg CO₂ — more than using it for 2 years
Positive Environmental Impacts of Technology
🌱 Technology helping the environment
- Smart energy grids: IoT sensors optimise electricity distribution, reducing waste and integrating renewable energy sources more efficiently
- Remote working: reduces commuting and business travel emissions
- Precision agriculture: sensors and data analytics reduce pesticide, water, and fertiliser use; drones monitor crop health
- Paperless systems: digital documents reduce paper, printing, and postage footprints
- Electric vehicle development: computing is enabling EV battery management, charging infrastructure, and autonomous driving
- Climate modelling: supercomputers model climate systems to improve predictions and inform policy
What Individuals Can Do
📱Keep devices longer — avoid upgrading every year; repair before replacing
♻️Use certified e-waste recyclers or donate working devices to charities
☁️Choose cloud providers that run on renewable energy
📴Power down devices not in use; enable energy saving modes
🛒Buy refurbished or second-hand electronics to extend device lifespans
📧Avoid sending unnecessary large emails; unsubscribe from unused mailing lists
Exam tip: Know both negative impacts (e-waste, energy consumption, carbon footprint, raw material extraction) AND positive impacts (smart grids, precision agriculture, paperless systems, remote work). AQA often asks "evaluate the environmental impact of digital technology" — you need both sides and a conclusion. Know the term "planned obsolescence" — it's a favourite.
⚠️ Common Mistakes
- Only mentioning e-waste — energy consumption and manufacturing footprint are equally important environmental impacts.
- Ignoring positive impacts — technology also helps tackle environmental problems. A balanced answer scores more.
- Confusing "carbon neutral" with "zero emissions" — carbon neutral means any emissions are offset (e.g. by planting trees or buying credits), not that no emissions are produced.