Moral and ethical issues: environmental effects, censorship, monitoring behaviour and analysing personal information

Overview

Recall questions on the environmental costs and benefits of computing, replacing school laptops, internet censorship and filtering, monitoring behaviour at work and elsewhere, and analysing personal information, re-identification, filter bubbles and loyalty card data.

What you need to know

Answer each one before you reveal it.

Questions from this subtopic

Each one with the model answer underneath.

1. 1

   What environmental costs are attributed to computing?

   Model answer

   Manufacture, which for many devices dominates their lifetime footprint: extraction and refining of metals including copper, gold, cobalt, tantalum and rare earth elements, with associated land damage, water use and, in some supply chains, documented labour abuses; and semiconductor fabrication, which is energy-intensive and uses very large volumes of ultrapure water and some potent greenhouse gases. Operation: the electricity used by devices, networks and data centres. The International Energy Agency has estimated data centre electricity use in the region of one and a half per cent of world demand, with cooling a large share of it, and has projected substantial growth driven by artificial intelligence workloads. Cryptocurrency mining is a separate and disputed line item; estimates for Bitcoin from the Cambridge Centre for Alternative Finance have ranged over roughly a hundred to nearly two hundred terawatt hours a year depending on assumptions about hardware mix. Disposal: the United Nations Global E-waste Monitor reported around sixty-two million tonnes of electronic waste generated in 2022, of which only about twenty-two per cent was documented as formally collected and recycled, with much of the rest informally processed in ways that release lead, mercury and dioxins. Short replacement cycles, driven by fashion, non-replaceable batteries, glued construction and withdrawal of software support, multiply all of these.

2. 2

   Make the environmental case in favour of computing.

   Model answer

   Substitution: video conferencing and remote working displace journeys, and a document delivered electronically replaces paper, printing, and physical distribution, though the size of the net saving depends on whether the journey is genuinely avoided or merely shifted. Optimisation: route planning cuts fuel in freight and delivery; smart grids match supply to demand and make intermittent renewable generation manageable; building management systems cut heating and lighting waste; precision agriculture targets water, fertiliser and pesticide to the square metre rather than the field. Dematerialisation: one device replaces a camera, a music collection, a satnav, a newspaper and a stack of maps, and streaming removes the manufacture and shipping of discs. Modelling: climate science, weather forecasting and materials research for batteries and catalysts all depend on large-scale computation, so some of the energy is spent on the problem itself. Monitoring: satellite and sensor data expose deforestation, methane leaks and illegal fishing that would otherwise go unrecorded. The honest caveat is the rebound effect, sometimes called the Jevons paradox: making an activity cheaper and more efficient tends to increase how much of it is done, so efficiency gains are frequently partly or wholly eaten by extra consumption, and cheap streaming and cheap cloud storage are good examples.

3. 3

   A school is deciding whether to replace working five-year-old laptops with new ones. Discuss the decision in environmental terms.

   Model answer

   The central fact is that for a laptop the embodied carbon of manufacture is typically a large share, often the majority, of lifetime emissions, so the greenest machine is usually the one already built. Extending the life of a working device avoids a new manufacturing footprint entirely and defers a disposal problem. Arguments for replacing anyway: newer machines may draw less power in use, though for a laptop that saving is small against manufacturing; if the old devices no longer receive security updates they may be unsafe to keep on the network, and unsupported hardware is a genuine constraint rather than an excuse; and if the machines are too slow to use, the environmental question is moot because they will simply be abandoned. Middle courses that a good answer should reach for: upgrade memory and replace the storage drive and battery to extend life; install a lighter operating system, which is a common route to keeping older hardware secure and usable; retain the fleet for a further two or three years and stagger replacement so the load is spread; and if replacing, insist on refurbished machines, take-back arrangements and repairability rather than sending the old ones to landfill. The decision also has an equity dimension: money spent on unnecessary hardware is money not spent on teaching.

4. 4

   What arguments are made in favour of restricting what can be published or accessed on the internet?

   Model answer

   Protection of children from sexual abuse material, and from content promoting suicide, self-harm and eating disorders, which was the driving argument behind recent United Kingdom legislation. Prevention of terrorism, where the concern is both recruitment material and operational instruction. Public safety and public health, where deliberate falsehoods about, say, vaccines or an evacuation route can cause direct harm. Protection of reputation and privacy, since defamation and the non-consensual publication of intimate images are already unlawful offline. Enforcement of intellectual property, which in the United Kingdom underpins court orders requiring internet service providers to block sites dedicated to infringement. National security. And the argument from consistency: broadcast, print and film have long been regulated, so an argument is needed for why the same content becomes acceptable when delivered over a different transport. Under the European Convention on Human Rights, freedom of expression is a qualified right, so restrictions are permitted where they are prescribed by law and necessary in a democratic society for aims including national security, public safety, the prevention of crime, the protection of health or morals, and the protection of the rights of others.

5. 5

   What are the arguments against internet censorship, and what practical problems do filtering schemes run into?

   Model answer

   Arguments of principle: free expression has value beyond the speaker, since a public able to hear unpopular claims is a precondition for correcting official error; the definition of harmful is contested and shifts, so a power created for one purpose is available to a later government with different views; and once infrastructure for blocking exists, it is much cheaper to extend than to build. Practical problems: over-blocking, where filters catch sex education, addiction support and domestic violence resources along with their targets, harming exactly the vulnerable users the scheme was meant to protect; under-blocking, since determined users route around filters with virtual private networks, proxies and Tor, so the burden falls hardest on the least technical; opacity, because block lists are often not published and there may be no way to learn you have been listed or to appeal; jurisdiction, since a site blocked in one country remains available from another; and delegation, because the practical arbiter becomes a private platform's moderation team or an industry body rather than a court. There is a separate line of criticism about chilling effects, where the mere possibility of blocking causes cautious self-censorship. Defenders reply that no measure needs to be perfect to be worthwhile, and that reducing the casual availability of harmful material has value even if the determined can evade it.

6. 6

   Compare state-led filtering with moderation by a private platform. Which raises the more serious concerns?

   Model answer

   State filtering is backed by force of law, applies to everyone in the jurisdiction and can criminalise the speaker, which makes it the more severe interference; but it is at least in principle constrained by legislation, subject to judicial review, and open to challenge under human rights law, so its limits are public and contestable. Platform moderation is not the state and a platform is entitled to set its own terms, and users can in theory leave; but a handful of platforms are where most public conversation now happens, the rules are written and changed unilaterally, enforcement is largely automated and error-prone at the volumes involved, the appeal route is often a form with no human at the end of it, and there is no equivalent of a court. Two further complications. First, the two interact: governments that would face legal obstacles to censoring directly can apply informal pressure on platforms, which is sometimes called jawboning, achieving the same result without the safeguards. Second, statutory duties of care push platforms towards over-removal, because the penalty for leaving harmful content up is far greater than the penalty for taking lawful content down. Which is more serious is genuinely disputed, and the answer depends on whether you weigh the severity of the sanction or the breadth of everyday reach.

Other ways to work this

The same content, four other ways.

What is inside Levo

Six ways to work the same material.