Main Notes

Why intelligent systems have an environmental impact

Intelligent systems can be useful because they analyse data, identify patterns and make decisions quickly. They appear in recommendation systems, voice assistants, traffic control, fraud detection, medical support tools, smart meters, factory robots and many other services. However, these systems do not run by magic. They need hardware, electricity, networks, storage, cooling and replacement devices. A good Higher answer should connect the visible service to the hidden computing resources behind it.

The environmental impact begins before the system is even used. Servers, phones, sensors and network equipment require raw materials to be mined, processed and transported. Manufacturing electronic components uses energy and water. If devices are replaced frequently, the impact of manufacture is spread over a shorter lifetime, which increases waste and demand for new materials.

Energy use: training, running and cooling

Some intelligent systems are trained using large sets of data. Training means adjusting the system so it can recognise patterns or make predictions. This can involve many calculations and may require powerful processors or graphics processors running for long periods. After training, the system still uses energy when it is used by people. Each prediction, recommendation, search, route calculation or automated decision may involve data being sent over a network, processed on a server and returned to a device.

Data centres are especially important. A data centre does not only power the servers. It also needs cooling because active processors produce heat. If the electricity comes from fossil fuels, the carbon footprint is higher. If the electricity comes from renewable sources, or if the data centre is designed to use cooling efficiently, the impact can be reduced. In an exam answer, avoid saying simply "AI is bad for the environment". The stronger answer explains which resources are being used and what conditions make the impact larger or smaller.

Positive environmental uses

Intelligent systems can also reduce environmental impact when they are designed and used well. The key idea is trade-off: a system may use energy itself but still reduce overall energy use elsewhere. The Higher course names three specific categories where this trade-off is examined — heating systems, traffic control, and car management systems. Each works the same way: sensors gather data, software makes a decision, and the environment benefits by cutting wasted energy or fuel.

Delivery route optimisation can also reduce fuel use by planning efficient routes, smart grids can balance electricity supply and demand to make better use of renewable energy, and monitoring systems can detect water leaks, pollution, equipment faults or crop problems earlier than manual checks.

Car management systems

Car management systems use sensors and an on-board computer — the Engine Control Unit (ECU) — to reduce the fuel a vehicle burns and the emissions it produces, without requiring the driver to change their behaviour.

Start/Stop technology automatically switches the engine off when the car is stationary, such as at traffic lights or in queued traffic, rather than leaving it idling. Sensors on the clutch or brake detect when the driver is ready to move off, and the engine restarts within milliseconds. A stationary, idling engine burns fuel while producing no useful motion, so cutting the engine during these periods reduces wasted fuel and the CO₂ produced by burning it.

The ECU continuously monitors engine sensors — including temperature, air quality and the fuel/air mixture — and adjusts fuel delivery in real time to keep combustion as efficient as possible. A poorly tuned fuel/air mixture wastes fuel and increases emissions; constant ECU monitoring keeps the engine closer to its optimum setting as driving conditions change.

Hybrid vehicles combine a combustion engine with an electric motor and battery. At low speeds the car can run on electric power alone, producing no exhaust emissions at all. Regenerative braking captures kinetic energy that would otherwise be lost as heat when braking, and uses it to recharge the battery instead of drawing that energy from fuel. Together, these features mean a hybrid vehicle typically burns less fuel overall than an equivalent petrol- or diesel-only vehicle.

The exam skill is the same for all three categories: name the specific technology, describe how sensors and software make a decision, and state the environmental benefit that results — rather than a vague claim such as "the car uses less fuel."

A system may use energy itself but still reduce overall energy use elsewhere. For example, a building management system uses sensors, networking and processing, but it might save more energy by controlling heating, lighting and ventilation. The answer depends on scale, lifetime and whether the system actually changes behaviour. A badly designed system might collect unnecessary data, run inefficiently, encourage more consumption or be replaced too often.

How organisations can reduce the impact

Organisations can reduce impact by using renewable electricity, efficient data centres, low-power hardware and efficient algorithms. They can avoid collecting data that is not needed, because unnecessary data must still be transmitted, stored, backed up and protected. They can extend device lifetime through repair, reuse and software updates. At the end of life, hardware should be recycled responsibly so useful materials can be recovered and hazardous materials are handled safely.

For Higher Computing Science, the most important skill is to write a precise, balanced explanation. Name the intelligent system, identify what it uses, describe the environmental cost, then consider whether it creates environmental savings. Mentioning both sides is usually stronger than giving a one-sided opinion.