Artificial intelligence is increasingly used to make decisions about hiring and lending.
A full IELTS Writing Task 2 answer to this question, with a paragraph plan, ideas you can use, and the vocabulary that fits this topic.
The question
You should spend about 40 minutes on this task.
Artificial intelligence is increasingly used to make decisions about hiring and lending.
What are the advantages and disadvantages of this?
Give reasons for your answer and include any relevant examples from your own knowledge or experience.
Write at least 250 words.
How to read this question
The strongest organising idea is that these systems learn from past decisions, so they inherit whatever bias produced those decisions — but express it consistently and at scale, which makes it simultaneously harder to see and easier to audit than human prejudice. Holding both halves of that paradox is what separates a thoughtful answer from a list of hopes and fears.
Paragraph plan
Introduction — Paraphrase, then signpost: real gains in consistency and access; real risks in inherited bias, opacity and the scale at which errors replicate.
Body 1 — Advantages — decisions become consistent and auditable, and lending can reach people whom conventional credit scoring never assessed at all.
Body 2 — Disadvantages — training data encodes past discrimination, proxies reintroduce protected characteristics, and an unexplainable refusal cannot be appealed.
Conclusion — The technology is neither fair nor unfair inherently; what decides is whether the system can be inspected and challenged.
Model essay
Band 8.5–9.0 model essay
318 words
Artificial intelligence is progressively being deployed to determine who is hired for a job or granted a loan. Although this trend can produce some merits, it can also lead to some issues.
On the one hand, this development carries two notable benefits. First, AI systems can process vast quantities of applications far more rapidly than human reviewers, screening thousands of resumes or loan requests within minutes rather than weeks. This accelerates recruitment and lending cycles considerably, benefiting both companies seeking to fill vacancies quickly and applicants awaiting a decision. Second, algorithmic assessment can apply identical criteria to every applicant without the fatigue, bias or mood fluctuations that affect human evaluators. To clarify, a human recruiter reviewing hundreds of applications in a single day may unconsciously apply stricter standards toward the end of the process, whereas software evaluates each case according to the same fixed parameters throughout.
On the other hand, this trend also entails two significant drawbacks. First, AI models are trained on historical data, which can embed existing societal biases into automated decisions. For instance, if past hiring or lending records disproportionately favoured certain demographic groups, the algorithm may learn to replicate this pattern, thereby perpetuating discrimination under the guise of technological neutrality, even when developers never intended such an outcome. Second, decisions generated by AI often lack transparency, leaving rejected applicants with little explanation or means of appeal. Since many systems function as opaque black boxes, even the companies deploying them may struggle to clarify why a particular candidate was declined, which can erode public trust in both employers and financial institutions, especially among applicants who suspect that an unseen error has cost them a genuine opportunity.
To conclude, the use of artificial intelligence in hiring and lending decisions can accelerate processing and apply consistent evaluation criteria, yet it also risks perpetuating embedded bias and reducing transparency and accountability in decisions that profoundly affect people's livelihoods.
Created by a Band 8.5–9.0 IELTS Writing achiever — Premium. Sign up free to read the rest of this essay and unlock every model on the site.
This is an advantages and disadvantages question. The essay above shows you what a strong answer looks like when it is finished. What it cannot show you is how to get there from a blank page in forty minutes.
That is what our Writing Study Library is for: the structure we teach for this exact question type, the paragraph pattern that goes with it, and the sentence openers for each stage — so the essay is planned before you start writing rather than assembled as you go.
Introduction: Paraphrase AI's growing role in hiring and lending decisions and preview that the essay will present both its merits and its issues.
Body 1: Give two advantages: AI can screen huge volumes of applications far faster than humans, speeding up recruitment and lending, and it applies identical criteria to everyone without the fatigue or mood swings that affect human evaluators.
Body 2: Give two disadvantages: models trained on historical data can embed and repeat past discrimination, and opaque 'black box' decisions leave rejected applicants without explanation or a means to appeal.
Conclusion: Summarise neutrally that AI speeds up and standardises hiring and lending decisions, while also risking embedded bias and reduced transparency in decisions that deeply affect people's lives.
Word or phrase
Meaning
Used in a sentence
algorithmic
relating to a set of rules followed by a computer to perform a task
algorithmic assessment can apply identical criteria to every applicant
screening
the process of evaluating candidates or applications to select suitable ones
screening thousands of resumes or loan requests within minutes rather than weeks
fluctuations
irregular rises and falls in something
without the fatigue, bias or mood fluctuations that affect human evaluators
embed
to fix firmly and deeply within something
AI models are trained on historical data, which can embed existing societal biases into automated decisions
disproportionately
to an extent that is too large or small in relation to something else
if past hiring or lending records disproportionately favoured certain demographic groups
perpetuating
causing something to continue indefinitely
thereby perpetuating discrimination under the guise of technological neutrality
opaque
not transparent; difficult to understand or see through
many systems function as opaque black boxes
erode
to gradually wear away or diminish
which can erode public trust in both employers and financial institutions
Created by a Band 8.5–9.0 IELTS Writing achiever — Premium. Sign up free to see the ideas and vocabulary for this topic and unlock every model on the site.
Answers checked here may be published on this page, without your name, to help other learners. Nothing else about you is stored.
Common questions
Should I use technical terms for a topic like this?
A few, defined implicitly by how you use them — "training data", "proxy" — help precision. What you should avoid is terminology you cannot explain in the same sentence; Lexical Resource rewards accurate use, not unfamiliar vocabulary on its own.
Is it a strong move to point out a paradox?
Yes, when it is genuine. Noting that algorithmic bias is harder to see but easier to prove than human bias shows you have thought past the obvious position, and it gives your conclusion something substantive to resolve.
Can both sides come from the same cause?
That is often the most elegant structure available. Deriving the advantages and disadvantages from one property — learning from past decisions — makes the essay cohere far better than two unrelated lists, and it demonstrates control of the topic.
