GP essay question guide
Should governments regulate artificial intelligence?
Define the harm and the type of AI before judging regulation. The strongest answer compares risk-based rules with standards, transparency duties, and support for innovation.
Unpack this question first: 'Should' asks for a judgement, not a description. Define 'regulate' (binding law, voluntary standards or both) and narrow 'artificial intelligence' to the uses you will discuss, such as hiring tools or chatbots. Lesson: Reading an essay question: terms and scope
Use this question in writing - Sign in, check the question, then write your response.
Three argument directions
Protect the public
Rules can require testing, accountability, and remedies where AI can cause serious harm.
Qualification: Broad rules can become obsolete quickly or burden low-risk uses.
Support trust and adoption
Clear duties can make firms and citizens more willing to use AI in important settings.
Qualification: Compliance labels do not prove that enforcement is capable or independent.
Keep room for innovation
Flexible standards and regulatory sandboxes can reduce risk without fixing one technical design in law.
Qualification: Voluntary or flexible systems can fail when commercial incentives reward speed over safety.
A model paragraph
Our own paragraph for this question. Copy its moves, not its words.
Governments should regulate artificial intelligence where its decisions can seriously harm people, because a firm racing to release a product has little reason to test it for rare but costly failures. A hiring tool that quietly screens out older applicants harms people who may never learn why they were rejected, so they cannot complain. Binding rules can require testing before use and give those affected a way to challenge the result. The European Union's AI Act takes this approach: it sorts uses by risk and puts the strictest duties on high-risk uses such as recruitment. Yet regulation is not always the best tool. For low-risk uses such as spam filters, heavy rules mostly add cost, and Singapore has preferred voluntary frameworks and testing tools that firms can adopt as the technology changes. Regulation is justified in proportion to the harm: strict where jobs and lives are at stake, light elsewhere.
How it is built
- Stand: regulate where harm is serious, not everywhere.
- Reason and case: a hiring tool that hides its bias; the EU's risk-based rules.
- Limit and judgement: low-risk uses need lighter tools, so match the rule to the harm.
Six examples to compare
Singapore's Agentic AI Governance Framework
Singapore published practical guardrails for AI systems that can plan and act, while keeping humans accountable.
Possible essay use
governments can manage the risks of new technology with targeted guardrails rather than bans, keeping humans accountable for what AI systems do
Limitation
This is a governance framework rather than evidence that every organisation follows the guidance or that regulation has eliminated AI harm.
Case or source date: 2026-01-22
Council of Europe Opens the First AI Treaty for Signature
The Council of Europe opened a treaty intended to align AI activities with human rights, democracy, and the rule of law.
Possible essay use
The Council of Europe opened a treaty intended to align AI activities with human rights, democracy, and the rule of law.
Limitation
The convention's influence depends on signatures, ratification, domestic measures, and how broadly exceptions are used.
Case or source date: 2024-09-05
Source: Council of Europe
UN General Assembly Adopts Its First Global AI Resolution
The UN General Assembly adopted a consensus resolution promoting safe, secure, and trustworthy AI consistent with human rights and sustainable development.
Possible essay use
global agreement on AI principles is possible, but non-binding resolutions rely on governments and firms choosing to follow them
Limitation
A non-binding resolution can establish principles but has limited direct enforcement over governments or firms.
Case or source date: 2024-03-21
Source: United Nations
Singapore Updates Its National AI Strategy
Singapore's National AI Strategy 2.0 sets out plans for AI talent, industry, research, infrastructure, and a trusted environment.
Possible essay use
governments can actively steer AI development by investing in talent, research and infrastructure, rather than leaving it entirely to markets
Limitation
Strategic investment can build capability but may misallocate resources or outpace accountability and workforce adaptation.
Case or source date: 2023-12-04
Source: Smart Nation Singapore
Governing AI: Ethics, Rules & Our Future
Debating ethical rules and global laws (like the EU AI Act) for advanced AI to ensure safety and align with human values.
Possible essay use
The EU AI Act entered into force on 1 August 2024, and its bans on unacceptable-risk AI have applied since 2 February 2025.
Limitation
The delay was a provisional agreement still needing formal adoption when this record was checked, and the Act's real effect depends on enforcement that had barely begun.
Case or source date: 2024-08-01
Source: European Commission
AI Sees All: Is Your Data Safe?
Explores how Big Data and AI erode personal privacy, fueling concerns about corporate and state surveillance. What are the safeguards?
Possible essay use
On 3 September 2024 the Dutch Data Protection Authority fined Clearview AI EUR 30.5 million for building an illegal database of faces scraped without consent.
Limitation
A fine shows a regulator's judgement, not whether it can be collected; Clearview disputes EU jurisdiction, so the case also shows the limits of national privacy law against foreign firms.
Case or source date: 2024-09-03
Source: Hunton Andrews Kurth (Privacy and Cybersecurity Law Blog)