AI Generated Exam Paper

A Level H1 General Paper Practice Paper 1

Free AI-Generated Owl Alpha A Level H1 General Paper Practice Paper 1 practice paper with questions and answers for Singapore students. This page is rendered as a direct URL so the questions and answers can be discovered without pressing in-page buttons.

These static practice materials are generated from the site's syllabus and paper-generation workflow, with source and model context shown so students and parents can evaluate the material before use.

A Level H1 General Paper AI Generated Generated by Owl Alpha Updated 2026-06-07

Back to Subject Browse Free Exam Papers

Questions

TuitionGoWhere Practice Paper - General Paper H1 A-Level

TuitionGoWhere Practice Paper (AI)

Subject: General Paper H1 Level: A-Level Paper: Practice Paper — Paper 2: Comprehension Duration: 1 hour 30 minutes Total Marks: 50 Name: ___________________________ Class: ___________________________ Date: ___________________________

Instructions

Answer all questions.
Read the passage carefully before attempting the questions.
For short-answer questions, use your own words as far as possible unless otherwise stated.
For the summary question, write your answer in continuous prose. Do not exceed the stated word limit.
For the application question, relate your answer to the context given and support your response with evidence or examples.
Marks are indicated in brackets [ ] at the end of each question or part-question.
The total marks for this paper is 50.

Section A: Short-Answer Questions [35 marks]

Read the passage below and answer Questions 1–15.

The Rise of Artificial Intelligence in Modern Healthcare

The integration of artificial intelligence (AI) into healthcare has accelerated at a pace that few could have anticipated a decade ago. From diagnostic imaging to drug discovery, AI systems are now capable of performing tasks that were once the exclusive domain of trained medical professionals. Proponents argue that this technological revolution promises to democratise healthcare, making high-quality diagnosis and treatment accessible to populations that have long been underserved. Yet, as with all transformative technologies, the rapid adoption of AI in medicine raises profound ethical, practical, and societal questions that demand careful scrutiny.

At the heart of the AI healthcare revolution lies machine learning — a subset of AI in which algorithms learn patterns from vast datasets without being explicitly programmed. In radiology, for instance, deep learning models have demonstrated the ability to detect tumours in medical scans with accuracy that rivals, and in some cases surpasses, that of experienced radiologists. A landmark study published in Nature Medicine in 2023 found that an AI system correctly identified breast cancer in mammograms 94.5% of the time, compared to 88% for human specialists. Such results have led some commentators to predict that AI will eventually replace radiologists altogether. However, this view is overly simplistic. While AI excels at pattern recognition within narrow parameters, it lacks the holistic clinical judgement that human doctors bring to patient care — the ability to consider a patient's history, emotional state, and social circumstances when making treatment decisions.

The pharmaceutical industry has also embraced AI with considerable enthusiasm. Traditional drug development is notoriously slow and expensive; it takes an average of 12 years and costs approximately $2.6 billion to bring a single drug to market. AI-driven platforms can dramatically compress this timeline by predicting how different molecular compounds will interact with biological targets, thereby identifying promising drug candidates in a fraction of the time. In 2024, the first AI-designed drug entered Phase III clinical trials, marking a watershed moment for the industry. Nevertheless, critics caution that the hype surrounding AI in drug discovery may be premature. Many AI-identified compounds still fail in clinical trials because biological systems are far more complex than any algorithm can fully model. The human body, with its intricate web of biochemical interactions, remains a frontier that AI has only begun to explore.

Beyond diagnostics and drug development, AI is reshaping the doctor-patient relationship itself. Telemedicine platforms powered by AI chatbots can now triage patients, answer routine medical questions, and even monitor chronic conditions through wearable devices. In rural areas of developing countries, where access to doctors is severely limited, these tools have the potential to save lives. A pilot programme in sub-Saharan Africa used an AI-powered mobile application to screen for diabetic retinopathy — a leading cause of blindness — achieving a detection rate of 91%. For communities that might otherwise go entirely without screening, such innovations represent a lifeline.

Yet the proliferation of AI in healthcare is not without significant risks. Data privacy is perhaps the most pressing concern. AI systems require enormous quantities of patient data to function effectively, and the collection, storage, and use of this data raise serious questions about consent and confidentiality. In 2023, a major hospital chain in the United States was found to have shared anonymised patient records with a technology company without explicit patient consent, sparking a public outcry and a class-action lawsuit. The incident underscored a troubling reality: the regulatory frameworks governing health data have not kept pace with technological innovation.

Algorithmic bias presents another formidable challenge. If the datasets used to train AI systems are not representative of diverse populations, the resulting models may perform poorly for certain demographic groups. Research has shown that some AI diagnostic tools are significantly less accurate for patients with darker skin tones, because the training data was predominantly drawn from lighter-skinned populations. This is not merely a technical glitch; it is a matter of equity. When AI systems perpetuate or amplify existing health disparities, they risk deepening the very inequalities they were supposed to help resolve.

There is also the question of accountability. When an AI system makes an incorrect diagnosis that leads to patient harm, who bears responsibility? The hospital that deployed the system? The developers who designed the algorithm? The regulators who approved it? Current legal frameworks are ill-equipped to address these questions, and the lack of clarity creates a dangerous grey area. Some legal scholars have proposed the concept of "shared liability," in which responsibility is distributed among all stakeholders. Others argue that AI should be treated as a tool — like a stethoscope or a scalpel — and that the human practitioner who uses it must bear ultimate responsibility.

Perhaps the most fundamental question is whether the increasing reliance on AI will erode the humanistic dimensions of medicine. Healing has always been more than a technical exercise; it involves empathy, trust, and the deeply personal connection between doctor and patient. If patients come to view healthcare as a purely algorithmic process — a series of inputs and outputs devoid of human warmth — something essential may be lost. The challenge, then, is not to resist AI, but to integrate it in ways that enhance rather than diminish the human experience of care.

As we stand at this crossroads, it is clear that the future of AI in healthcare will be shaped not only by technological advances but by the choices we make as a society. The decisions we take today about regulation, equity, and the role of human judgement in medicine will determine whether AI becomes a force for genuine progress or a source of new and unforeseen problems.

Question 1

In paragraph 1, the author states that AI systems are "now capable of performing tasks that were once the exclusive domain of trained medical professionals." What does the phrase "exclusive domain" suggest about the role of medical professionals in the past? [2]

Question 2

In paragraph 1, the author uses the word "democratise" to describe the potential impact of AI on healthcare. Explain what the author means by this word in the context of the passage. [2]

Question 3

In paragraph 2, the author cites a study published in Nature Medicine in 2023. Why does the author include this specific piece of evidence? [2]

Question 4

In paragraph 2, the author writes that the view that AI will replace radiologists is "overly simplistic." What does the author mean by this phrase, and what evidence does the author provide to support this claim? [3]

Question 5

In paragraph 3, the author describes the first AI-designed drug entering Phase III clinical trials as a "watershed moment." Explain the author's use of this phrase. [2]

Question 6

In paragraph 3, the author states that "biological systems are far more complex than any algorithm can fully model." What does this suggest about the limitations of AI in drug discovery? [2]

Question 7

In paragraph 4, the author describes AI-powered screening tools in sub-Saharan Africa as "a lifeline." What does this metaphor suggest about the impact of AI in developing regions? [2]

Question 8

In paragraph 5, the author mentions a hospital chain that "shared anonymised patient records with a technology company without explicit patient consent." What concern does this incident illustrate? [2]

Question 9

In paragraph 6, the author states that algorithmic bias "is not merely a technical glitch; it is a matter of equity." Explain the author's use of the word "equity" in this context. [2]

Question 10

In paragraph 7, the author refers to "a dangerous grey area" in relation to accountability for AI errors. What does the author mean by this phrase? [2]

Question 11

In paragraph 8, the author asks whether AI will "erode the humanistic dimensions of medicine." What does the author mean by "humanistic dimensions"? [2]

Question 12

In the final paragraph, the author writes that "the future of AI in healthcare will be shaped not only by technological advances but by the choices we make as a society." What two factors does the author suggest will determine the future of AI in healthcare? [2]

Question 13

Throughout the passage, the author presents both the benefits and risks of AI in healthcare. Identify one benefit and one risk mentioned in the passage, and briefly explain how the author develops each point. [4]

Benefit: _______________________________________________________________________

Risk: ________________________________________________________________________

Question 14

How does the author structure the passage to present a balanced argument about AI in healthcare? Refer to specific paragraphs in your answer. [4]

Question 15

The author begins paragraph 1 with a general statement about the pace of AI integration and ends the passage with a forward-looking statement about societal choices. How does this framing contribute to the overall purpose of the passage? [3]

Section B: Summary Question [8 marks]

Question 16

Summarise the risks and challenges associated with the use of AI in healthcare, as outlined in the passage.

Use your own words as far as possible. Write your summary in no more than 120 words.

Section C: Application Question [7 marks]

Question 17

The passage discusses the use of AI in healthcare. Consider the following scenario:

Your country's Ministry of Health is considering a proposal to deploy AI diagnostic systems in all public hospitals within the next five years. The proposal argues that AI will reduce waiting times, lower costs, and improve diagnostic accuracy. However, some healthcare workers and patient advocacy groups have raised concerns about data privacy, job displacement, and the loss of personal interaction between doctors and patients.

As a policy advisor, write a response to the Ministry of Health in which you:

(a) Explain two potential benefits of deploying AI diagnostic systems in public hospitals, using evidence from the passage. [3]

(b) Discuss two concerns that should be addressed before the proposal is implemented, using evidence from the passage. [4]

End of Paper

Section A Total: 35 marks Section B Total: 8 marks Section C Total: 7 marks Grand Total: 50 marks

Answers

TuitionGoWhere Practice Paper — General Paper H1 A-Level

Answer Key and Marking Scheme

Paper: Practice Paper — Paper 2: Comprehension Total Marks: 50

Section A: Short-Answer Questions [35 marks]

Question 1 [2 marks]

Question: In paragraph 1, the author states that AI systems are "now capable of performing tasks that were once the exclusive domain of trained medical professionals." What does the phrase "exclusive domain" suggest about the role of medical professionals in the past?

Answer: The phrase "exclusive domain" suggests that in the past, only trained medical professionals had the knowledge, skills, and authority to perform certain healthcare tasks. These tasks were not accessible to machines or untrained individuals. The word "exclusive" implies that this was a privileged, restricted area of practice belonging solely to qualified humans.

Marking Scheme:

1 mark for identifying that the tasks were performed only by trained professionals.
1 mark for explaining the implication of restriction/privilege (i.e., not accessible to others/machines).

Common Mistakes:

Students may simply lift "trained medical professionals" without explaining what "exclusive domain" implies.
Students may confuse "exclusive" with "extensive" or fail to convey the sense of restriction.

Question 2 [2 marks]

Question: In paragraph 1, the author uses the word "democratise" to describe the potential impact of AI on healthcare. Explain what the author means by this word in the context of the passage.

Answer: In this context, "democratise" means to make high-quality healthcare accessible to all people, regardless of their geographic location, socioeconomic status, or proximity to medical facilities. The author suggests that AI can level the playing field by bringing expert-level diagnosis and treatment to underserved populations who previously had limited access.

Marking Scheme:

1 mark for explaining the general meaning: making something available to everyone.
1 mark for contextualising it to healthcare (i.e., access for underserved/remote populations).

Common Mistakes:

Students may give a political definition of "democratise" unrelated to healthcare access.
Students may fail to connect the word to the idea of accessibility for underserved populations mentioned in the passage.

Question 3 [2 marks]

Question: In paragraph 2, the author cites a study published in Nature Medicine in 2023. Why does the author include this specific piece of evidence?

Answer: The author includes this evidence to provide concrete, credible data supporting the claim that AI can match or exceed human performance in specific diagnostic tasks. By citing a reputable journal (Nature Medicine) and specific accuracy figures (94.5% vs. 88%), the author strengthens the argument that AI has genuine potential in healthcare and is not merely speculative.

Marking Scheme:

1 mark for identifying that the evidence supports the claim about AI's diagnostic accuracy.
1 mark for explaining the effect of using a credible source/specific data (i.e., adds credibility, makes the argument persuasive).

Common Mistakes:

Students may simply restate the statistics without explaining why the author included them.
Students may fail to mention the persuasive/credibility function of citing a named journal.

Question 4 [3 marks]

Question: In paragraph 2, the author writes that the view that AI will replace radiologists is "overly simplistic." What does the author mean by this phrase, and what evidence does the author provide to support this claim?

Answer: The phrase "overly simplistic" means that the view is too narrow and fails to consider the full complexity of medical practice. The author argues that while AI excels at pattern recognition in specific tasks, it cannot replicate the holistic clinical judgement of human doctors. The evidence provided is that human doctors consider a patient's full history, emotional state, and social circumstances — factors that AI cannot account for.

Marking Scheme:

1 mark for explaining "overly simplistic" (too narrow / ignores complexity).
1 mark for identifying the limitation of AI (lacks holistic clinical judgement).
1 mark for identifying what human doctors do that AI cannot (consider patient history, emotions, social circumstances).

Common Mistakes:

Students may only explain the phrase without providing the supporting evidence.
Students may describe what AI can do rather than what it cannot do compared to humans.

Question 5 [2 marks]

Question: In paragraph 3, the author describes the first AI-designed drug entering Phase III clinical trials as a "watershed moment." Explain the author's use of this phrase.

Answer: A "watershed moment" refers to a turning point or pivotal event that marks the beginning of a new era. By using this phrase, the author emphasises that the entry of an AI-designed drug into Phase III trials represents a historic milestone for the pharmaceutical industry — the first time AI has progressed a drug to such an advanced stage of development, signalling a fundamental shift in how drugs are discovered.

Marking Scheme:

1 mark for explaining the meaning of "watershed moment" (turning point / pivotal milestone).
1 mark for contextualising it to the significance for the pharmaceutical industry.

Common Mistakes:

Students may not know the meaning of "watershed" and guess incorrectly.
Students may explain the meaning but fail to connect it to the context of drug development.

Question 6 [2 marks]

Question: In paragraph 3, the author states that "biological systems are far more complex than any algorithm can fully model." What does this suggest about the limitations of AI in drug discovery?

Answer: This suggests that AI, no matter how advanced, has inherent limitations in drug discovery because it cannot fully capture the complexity of the human body's biochemical interactions. While AI can identify promising drug candidates, it may miss critical factors that only emerge in real biological systems, which is why many AI-identified compounds still fail in clinical trials.

Marking Scheme:

1 mark for identifying that AI cannot fully model/comprehend biological complexity.
1 mark for explaining the consequence (compounds may fail in trials; AI has limits).

Common Mistakes:

Students may simply restate the quote without explaining its implication.
Students may fail to connect the limitation to the failure of compounds in clinical trials.

Question 7 [2 marks]

Question: In paragraph 4, the author describes AI-powered screening tools in sub-Saharan Africa as "a lifeline." What does this metaphor suggest about the impact of AI in developing regions?

Answer: The metaphor "a lifeline" suggests that AI-powered screening tools are essential and potentially life-saving for communities in developing regions. Just as a lifeline rescues someone from danger, these tools provide critical medical screening to populations that would otherwise have little or no access to such services, thereby preventing avoidable deaths and diseases like blindness from diabetic retinopathy.

Marking Scheme:

1 mark for explaining the metaphor (something essential / life-saving).
1 mark for connecting it to the context (underserved communities with no alternative access).

Common Mistakes:

Students may explain the metaphor generically without linking it to the healthcare context.
Students may not mention the idea of these communities having no alternative.

Question 8 [2 marks]

Question: In paragraph 5, the author mentions a hospital chain that "shared anonymised patient records with a technology company without explicit patient consent." What concern does this incident illustrate?

Answer: This incident illustrates the concern about data privacy and the lack of adequate regulatory frameworks governing the use of patient data. It shows that even when data is anonymised, the sharing of personal health information without patients' explicit consent violates their privacy rights and erodes trust in healthcare institutions.

Marking Scheme:

1 mark for identifying the concern about data privacy / patient consent.
1 mark for explaining the broader implication (regulatory gaps / erosion of trust).

Common Mistakes:

Students may only describe the incident without identifying the underlying concern.
Students may fail to mention the regulatory gap highlighted by the author.

Question 9 [2 marks]

Question: In paragraph 6, the author states that algorithmic bias "is not merely a technical glitch; it is a matter of equity." Explain the author's use of the word "equity" in this context.

Answer: In this context, "equity" refers to fairness and justice in healthcare outcomes across different demographic groups. The author is arguing that when AI systems perform less accurately for certain populations (e.g., those with darker skin tones), it is not just a technical problem but a fundamental issue of unfairness — it means that some groups receive inferior healthcare, which deepens existing social inequalities.

Marking Scheme:

1 mark for defining "equity" as fairness / justice across groups.
1 mark for connecting it to the consequence (deepening health disparities / unequal care).

Common Mistakes:

Students may confuse "equity" with "equality" without explaining the distinction in context.
Students may not connect the term to the specific issue of demographic bias in AI.

Question 10 [2 marks]

Question: In paragraph 7, the author refers to "a dangerous grey area" in relation to accountability for AI errors. What does the author mean by this phrase?

Answer: The phrase "a dangerous grey area" refers to the lack of clear legal and ethical frameworks for determining who is responsible when an AI system causes patient harm. Because it is unclear whether the hospital, the developers, or the regulators should be held accountable, this ambiguity creates a risky situation where patients may be harmed without any party being properly answerable.

Marking Scheme:

1 mark for identifying the lack of clarity in accountability / legal frameworks.
1 mark for explaining why it is "dangerous" (patients harmed without recourse; no one held responsible).

Common Mistakes:

Students may describe the grey area without explaining why it is dangerous.
Students may list who could be responsible without explaining the ambiguity.

Question 11 [2 marks]

Question: In paragraph 8, the author asks whether AI will "erode the humanistic dimensions of medicine." What does the author mean by "humanistic dimensions"?

Answer: The "humanistic dimensions" of medicine refer to the aspects of healthcare that go beyond technical diagnosis and treatment — namely, the empathy, trust, emotional support, and personal connection between doctor and patient. The author is concerned that over-reliance on AI could reduce healthcare to a purely mechanical process, stripping away the compassionate and relational elements that are central to healing.

Marking Scheme:

1 mark for identifying the non-technical aspects (empathy, trust, personal connection).
1 mark for explaining the concern (healthcare becoming mechanical / losing human warmth).

Common Mistakes:

Students may give a vague answer like "the human side" without specifying what that entails.
Students may fail to connect the term to the doctor-patient relationship.

Question 12 [2 marks]

Question: In the final paragraph, the author writes that "the future of AI in healthcare will be shaped not only by technological advances but by the choices we make as a society." What two factors does the author suggest will determine the future of AI in healthcare?

Answer: The two factors are: (1) technological advances (i.e., improvements in AI capabilities and systems), and (2) societal choices (i.e., decisions about regulation, equity, and the role of human judgement in medicine).

Marking Scheme:

1 mark for each factor identified (1 mark for technological advances, 1 mark for societal choices).

Common Mistakes:

Students may only identify one factor.
Students may paraphrase vaguely without clearly distinguishing the two factors.

Question 13 [4 marks]

Question: Throughout the passage, the author presents both the benefits and risks of AI in healthcare. Identify one benefit and one risk mentioned in the passage, and briefly explain how the author develops each point.

Answer:

Benefit (example): AI can improve diagnostic accuracy and accessibility. The author develops this by citing specific evidence — the Nature Medicine study showing AI's 94.5% accuracy in detecting breast cancer, and the sub-Saharan Africa pilot programme achieving a 91% detection rate for diabetic retinopathy. These concrete examples demonstrate AI's practical value.

Risk (example): Algorithmic bias can worsen health disparities. The author develops this by explaining that AI training datasets often underrepresent certain demographic groups (e.g., darker-skinned patients), leading to less accurate diagnoses for those populations. The author frames this not as a technical issue but as a matter of equity, elevating the significance of the risk.

Marking Scheme:

1 mark for identifying a valid benefit with brief explanation of how the author develops it.
1 mark for identifying a valid risk with brief explanation of how the author develops it.
1 mark for referencing specific evidence from the passage for the benefit.
1 mark for referencing specific evidence from the passage for the risk.

Acceptable answers include:

Benefits: improved diagnostics, faster drug development, increased access in rural/developing areas, reduced costs.
Risks: data privacy breaches, algorithmic bias, accountability gaps, loss of humanistic care.

Common Mistakes:

Students may identify a benefit or risk without explaining how the author develops the point.
Students may make general statements without referencing specific evidence from the passage.

Question 14 [4 marks]

Question: How does the author structure the passage to present a balanced argument about AI in healthcare? Refer to specific paragraphs in your answer.

Answer: The author structures the passage by alternating between the benefits and risks of AI in healthcare, creating a balanced and measured argument. Paragraphs 1–4 primarily present the benefits and potential of AI: paragraph 1 introduces the transformative potential, paragraph 2 discusses diagnostic accuracy, paragraph 3 covers drug development, and paragraph 4 highlights accessibility in developing regions. Paragraphs 5–8 then systematically address the risks and challenges: data privacy (paragraph 5), algorithmic bias (paragraph 6), accountability (paragraph 7), and the loss of humanistic care (paragraph 8). The passage concludes in paragraph 9 with a forward-looking statement that synthesises both sides, emphasising that the future depends on both technology and societal choices. This structure allows the reader to appreciate AI's potential while remaining aware of its significant challenges.

Marking Scheme:

1 mark for identifying the alternating/benefit-risk structure.
1 mark for referencing specific paragraphs that present benefits (any from paras 1–4).
1 mark for referencing specific paragraphs that present risks (any from paras 5–8).
1 mark for explaining how the conclusion synthesises both sides.

Common Mistakes:

Students may describe the structure without referencing specific paragraphs.
Students may only discuss benefits or only risks, failing to address the balance.

Question 15 [3 marks]

Question: The author begins paragraph 1 with a general statement about the pace of AI integration and ends the passage with a forward-looking statement about societal choices. How does this framing contribute to the overall purpose of the passage?

Answer: The opening statement about the rapid pace of AI integration immediately establishes the significance and urgency of the topic, capturing the reader's attention and signalling that this is a development that cannot be ignored. The closing statement about societal choices shifts the focus from technology alone to human agency, reinforcing the passage's purpose of encouraging critical reflection rather than passive acceptance. Together, this framing creates a narrative arc that moves from describing a technological phenomenon to urging readers to think carefully about how it should be governed. This contributes to the passage's overall purpose of presenting a balanced, thought-provoking analysis that neither celebrates nor condemns AI, but calls for informed and responsible decision-making.

Marking Scheme:

1 mark for explaining the effect of the opening (establishes significance/urgency).
1 mark for explaining the effect of the closing (shifts to human agency/critical reflection).
1 mark for explaining how the framing contributes to the overall purpose (balanced analysis, calls for responsible decision-making).

Common Mistakes:

Students may describe the content of the opening and closing without explaining their rhetorical effect.
Students may fail to connect the framing to the passage's overall purpose.

Section B: Summary Question [8 marks]

Question 16 [8 marks]

Question: Summarise the risks and challenges associated with the use of AI in healthcare, as outlined in the passage. Use your own words as far as possible. Write your summary in no more than 120 words.

Content Points (any 8 valid points for 1 mark each):

AI systems require vast amounts of patient data, raising concerns about data privacy and confidentiality.
Patient data may be shared without explicit consent, as seen in the case of a US hospital chain.
Existing regulatory frameworks have not kept pace with technological innovation, creating legal gaps.
Algorithmic bias occurs when training datasets are not representative of diverse populations.
AI diagnostic tools may be less accurate for certain demographic groups (e.g., darker-skinned patients).
This bias can perpetuate or worsen existing health disparities.
There is a lack of clarity about accountability when AI systems cause patient harm.
It is unclear whether hospitals, developers, or regulators should bear responsibility.
Over-reliance on AI may erode the empathetic, human connection between doctor and patient.
Healthcare could become a purely algorithmic process, losing the trust and warmth essential to healing.

Marking Scheme:

Content: Up to 8 marks — 1 mark for each valid point summarised in the student's own words (maximum 8 points from the list above).
Language: Integrated into content marks — points that are directly lifted from the passage should not be awarded marks. Students must demonstrate paraphrasing.
Word limit: Answers exceeding 120 words should be penalised. The summary should stop being marked once the word limit is exceeded.

Sample Summary (for reference):

The use of AI in healthcare presents several risks. The technology requires enormous quantities of patient data, raising privacy concerns, especially when data is shared without consent. Regulatory frameworks have failed to keep up with these developments. Additionally, algorithmic bias is a significant challenge: AI systems trained on unrepresentative data may be less accurate for certain groups, worsening health inequalities. Accountability is also unclear — when AI causes harm, it is uncertain who should be held responsible. Finally, the increasing use of AI may diminish the empathetic relationship between doctor and patient, reducing healthcare to a mechanical process devoid of human warmth. (98 words)

Common Mistakes:

Students may include benefits of AI rather than focusing only on risks and challenges.
Students may lift phrases directly from the passage instead of paraphrasing.
Students may exceed the 120-word limit.
Students may include their own opinions or examples not found in the passage.

Section C: Application Question [7 marks]

Question 17 [7 marks]

Question: As a policy advisor, write a response to the Ministry of Health in which you:

(a) Explain two potential benefits of deploying AI diagnostic systems in public hospitals, using evidence from the passage. [3 marks]

Answer:

Benefit 1: Improved diagnostic accuracy and efficiency. The passage cites a 2023 study in Nature Medicine showing that AI correctly identified breast cancer in mammograms 94.5% of the time, compared to 88% for human specialists. This evidence suggests that AI diagnostic systems could enhance the accuracy of diagnoses in public hospitals, potentially catching conditions that human doctors might miss and reducing misdiagnoses.

Benefit 2: Increased accessibility, particularly for underserved populations. The passage describes how an AI-powered mobile application in sub-Saharan Africa achieved a 91% detection rate for diabetic retinopathy, providing screening to communities that would otherwise have no access. Similarly, deploying AI in public hospitals could extend diagnostic capabilities to underserved areas within the country, reducing waiting times and improving equity of access.

Marking Scheme (part a):

1 mark for identifying Benefit 1 (improved accuracy/efficiency).
1 mark for supporting Benefit 1 with evidence from the passage.
1 mark for identifying Benefit 2 (increased accessibility) with evidence from the passage.

(b) Discuss two concerns that should be addressed before the proposal is implemented, using evidence from the passage. [4 marks]

Answer:

Concern 1: Data privacy and security. The passage highlights a case where a major US hospital chain shared anonymised patient records with a technology company without explicit patient consent, leading to public outrage and legal action. Before deploying AI diagnostic systems, the Ministry must establish robust data protection regulations to ensure that patient data is collected, stored, and used with informed consent and adequate security measures. Without such safeguards, public trust in the healthcare system could be severely damaged.

Concern 2: Algorithmic bias and equity. The passage notes that some AI diagnostic tools are significantly less accurate for patients with darker skin tones because their training data was predominantly drawn from lighter-skinned populations. If the AI systems deployed in public hospitals are trained on unrepresentative data, they could produce less accurate diagnoses for certain demographic groups, worsening existing health disparities. The Ministry must ensure that training datasets are diverse and representative of the country's population, and that AI systems are rigorously tested across all demographic groups before deployment.

Marking Scheme (part b):

1 mark for identifying Concern 1 (data privacy) with explanation.
1 mark for supporting Concern 1 with evidence from the passage.
1 mark for identifying Concern 2 (algorithmic bias) with explanation.
1 mark for supporting Concern 2 with evidence from the passage.

Additional marking notes for part (b):

Answers that discuss other valid concerns from the passage (e.g., accountability for AI errors, loss of humanistic care, job displacement) should also be accepted if supported with evidence and reasoning.
Students should demonstrate the ability to apply passage concepts to the specific scenario, not merely restate passage content.

Common Mistakes:

Students may discuss benefits or concerns without referencing evidence from the passage.
Students may provide generic answers that could apply to any technology, rather than specifically addressing AI in healthcare.
Students may introduce their own examples not found in the passage — while not penalised, the question specifically asks for evidence from the passage.

End of Answer Key

Section A Total: 35 marks Section B Total: 8 marks Section C Total: 7 marks Grand Total: 50 marks