IELTS Reading: Công nghệ ảnh hưởng chính sách công – Đề thi mẫu có đáp án chi tiết

Mở bài

Chủ đề về công nghệ và chính sách công (How Is Technology Influencing Public Policy?) đang trở thành một trong những đề tài “hot” xuất hiện thường xuyên trong IELTS Reading. Với sự phát triển vũ bão của trí tuệ nhân tạo, big data, và công nghệ số, các chính phủ trên toàn cầu đang phải điều chỉnh cách thức hoạch định chính sách. Đây là chủ đề liên ngành giữa công nghệ, chính trị, và xã hội – những lĩnh vực mà IELTS rất yêu thích.

Trong bài viết này, bạn sẽ nhận được một đề thi IELTS Reading hoàn chỉnh với 3 passages được thiết kế theo đúng chuẩn Cambridge IELTS, từ độ khó Easy đến Hard. Mỗi passage đi kèm với 40 câu hỏi đa dạng dạng bài, đáp án chi tiết với giải thích cụ thể, và bộ từ vựng quan trọng giúp bạn nâng cao vốn từ học thuật.

Đề thi này phù hợp cho học viên từ band 5.0 trở lên, giúp bạn làm quen với cấu trúc đề thi thật, rèn luyện kỹ năng skimming, scanning, và đặc biệt là khả năng paraphrase – yếu tố then chốt để đạt điểm cao IELTS Reading.

1. Hướng dẫn làm bài IELTS Reading

Tổng Quan Về IELTS Reading Test

IELTS Reading Test kéo dài 60 phút với 3 passages và tổng cộng 40 câu hỏi. Bạn cần tự quản lý thời gian hiệu quả vì không có thời gian chuyển đáp án riêng.

Phân bổ thời gian khuyến nghị:

• Passage 1: 15-17 phút (độ khó thấp, câu hỏi dễ tìm)
• Passage 2: 18-20 phút (độ khó trung bình, cần suy luận nhiều hơn)
• Passage 3: 23-25 phút (độ khó cao, từ vựng phức tạp, câu hỏi khó)

Mỗi câu trả lời đúng được 1 điểm, không bị trừ điểm khi sai. Điểm thô (raw score) từ 40 câu sẽ được quy đổi thành band score từ 1-9.

Các Dạng Câu Hỏi Trong Đề Này

Đề thi mẫu này bao gồm 7 dạng câu hỏi phổ biến nhất:

1. Multiple Choice – Trắc nghiệm nhiều lựa chọn
2. True/False/Not Given – Xác định thông tin đúng/sai/không có
3. Yes/No/Not Given – Xác định ý kiến tác giả
4. Matching Headings – Ghép tiêu đề với đoạn văn
5. Sentence Completion – Hoàn thành câu
6. Summary Completion – Hoàn thành tóm tắt
7. Short-answer Questions – Câu hỏi trả lời ngắn

---

2. IELTS Reading Practice Test

PASSAGE 1 – Digital Democracy: How Technology is Reshaping Public Participation

Độ khó: Easy (Band 5.0-6.5)

Thời gian đề xuất: 15-17 phút

The way citizens interact with their governments has undergone a dramatic transformation over the past two decades. Traditional methods of public consultation, such as town hall meetings and postal surveys, are increasingly being supplemented – and sometimes replaced – by digital platforms that promise greater accessibility and efficiency. This shift represents one of the most visible ways technology is influencing public policy, making governance potentially more responsive to citizens’ needs.

E-democracy, or electronic democracy, encompasses various forms of digital participation in political processes. At its most basic level, this includes government websites providing information about policies and services. More sophisticated applications allow citizens to submit feedback, participate in online polls, and even contribute directly to policy formulation. For instance, the city of Madrid launched its Decide Madrid platform in 2015, enabling residents to propose policy initiatives. If a proposal receives enough support from other citizens, it advances to a public vote. This direct democracy approach has engaged over 400,000 participants since its inception.

Social media platforms have become unexpected tools for policy influence. Government agencies now maintain active presences on Twitter, Facebook, and Instagram, not merely to broadcast information but to engage in two-way communication with constituents. During emergencies, these channels provide real-time updates and gather crowd-sourced information about developing situations. The 2011 earthquake in Japan demonstrated how Twitter could serve as an early warning system, with automated alerts reaching citizens faster than traditional broadcast media.

Công dân tham gia vào quá trình chính sách công thông qua nền tảng kỹ thuật số và e-democracy

However, the digitalization of public participation raises important questions about accessibility and representation. While technology enthusiasts celebrate these innovations, critics point out that not all citizens have equal access to digital tools. The digital divide – the gap between those with and without reliable internet access and digital literacy – means that online consultation processes may inadvertently exclude marginalized communities. Elderly populations, rural residents with poor connectivity, and low-income households without personal devices face barriers to participation that didn’t exist with traditional methods.

Moreover, the quality of online discourse often differs from face-to-face interactions. Anonymity can encourage more honest feedback but may also facilitate harassment and unconstructive criticism. Policymakers must develop strategies to moderate online discussions without censoring legitimate viewpoints, a challenging balance that continues to evolve. Some jurisdictions employ professional facilitators for online consultations, while others rely on community moderation models similar to those used by social media platforms.

Data analytics represents another dimension of technology’s influence on public policy. Governments can now analyze patterns in citizen feedback at scales previously impossible. Artificial intelligence tools can process thousands of survey responses, identifying common themes and concerns that might be missed in manual analysis. The city of Boston’s “Citizens Connect” app allows residents to report issues like potholes or broken streetlights. The accumulated data helps city planners identify infrastructure priorities and allocate resources more efficiently.

The transparency enabled by technology also transforms governance. Many governments now publish datasets online, allowing citizens, journalists, and researchers to examine public spending, service delivery statistics, and policy outcomes. This open data movement creates accountability mechanisms that can discourage corruption and inefficiency. When citizens can easily access information about how tax money is spent or track the progress of government projects, officials face greater pressure to demonstrate responsible stewardship of public resources.

Despite these advances, technology remains a tool rather than a solution in itself. Effective democratic participation requires more than just platforms – it demands political will to genuinely consider citizen input, institutional capacity to process feedback, and ongoing efforts to ensure inclusive access. As one policy expert noted, “Technology can amplify democracy, but it cannot create democracy where the commitment doesn’t exist.”

Questions 1-6

Do the following statements agree with the information given in Passage 1?

Write:

• TRUE if the statement agrees with the information
• FALSE if the statement contradicts the information
• NOT GIVEN if there is no information on this

1. Traditional consultation methods are completely being abandoned in favor of digital platforms.
2. The Decide Madrid platform requires proposals to gain sufficient citizen support before proceeding to a public vote.
3. Social media proved more effective than traditional media in providing earthquake warnings in Japan in 2011.
4. The digital divide affects certain demographic groups more than others.
5. Professional facilitators are the only method used to moderate online policy discussions.
6. Boston’s Citizens Connect app has eliminated all infrastructure problems in the city.

Questions 7-10

Complete the sentences below.

Choose NO MORE THAN TWO WORDS from the passage for each answer.

7. E-democracy includes various forms of __ in political processes.
8. Government agencies use social media for __ with their constituents, not just to share information.
9. Critics argue that online consultation may unintentionally exclude __.
10. The open data movement creates __ that help prevent corruption.

Questions 11-13

Choose the correct letter, A, B, C or D.

11. According to the passage, what is one benefit of data analytics in public policy?

• A) It eliminates the need for human analysis
• B) It can identify patterns in large volumes of citizen feedback
• C) It guarantees better policy outcomes
• D) It replaces traditional survey methods entirely

12. The passage suggests that anonymity in online discussions:

• A) Should always be prevented
• B) Has only negative consequences
• C) Has both advantages and disadvantages
• D) Is preferred by all participants

13. What does the passage conclude about technology’s role in democracy?

• A) Technology alone can create effective democracy
• B) Technology is unnecessary for democratic participation
• C) Technology can enhance democracy but requires other commitments
• D) Technology has failed to improve governance

---

PASSAGE 2 – Algorithmic Governance: When AI Meets Policy Making

Độ khó: Medium (Band 6.0-7.5)

Thời gian đề xuất: 18-20 phút

The integration of artificial intelligence into governmental decision-making processes represents a profound shift in how public policy is formulated and implemented. Unlike the digital platforms discussed in earlier contexts of citizen engagement, algorithmic governance involves using computational systems to directly influence or make policy decisions. This development raises fundamental questions about accountability, transparency, and the proper balance between human judgment and machine efficiency.

Predictive policing provides one of the most controversial examples of algorithm-driven policy. Police departments in cities ranging from Los Angeles to London employ software that analyzes crime data to identify locations where criminal activity is statistically more likely to occur. Advocates argue this approach enables more efficient resource allocation, directing patrols to areas of highest need. The PredPol system, used in over 60 jurisdictions, claims to predict crime with significantly greater accuracy than human analysts. However, critics raise concerns about self-fulfilling prophecies: increased police presence in predicted areas naturally leads to more arrests, which feeds back into the algorithm, potentially creating cycles of over-policing in certain neighborhoods.

The opaque nature of many AI systems compounds these concerns. Machine learning algorithms, particularly deep learning models, often function as “black boxes” – their decision-making processes are not easily interpretable even by their creators. When an algorithm recommends denying someone a business permit or targeting a neighborhood for increased surveillance, affected individuals may have no meaningful way to understand or challenge the reasoning. This opacity conflicts with fundamental principles of administrative law, which traditionally require that government decisions be explainable and subject to review.

Some jurisdictions are pioneering approaches to algorithmic accountability. New York City passed legislation in 2017 establishing a task force to study how automated decision systems are used by city agencies and to recommend procedures for public disclosure. The European Union’s General Data Protection Regulation (GDPR) includes provisions for a “right to explanation” when individuals are subject to automated decision-making with significant effects. However, implementing these principles proves challenging. Technical explanations of algorithmic functioning may be incomprehensible to non-experts, while simplified explanations may omit crucial details that would reveal biases or errors.

Trí tuệ nhân tạo và thuật toán ảnh hưởng đến quy trình ra quyết định chính sách công

Algorithmic bias presents another critical challenge. AI systems learn from historical data, and when that data reflects past discrimination, algorithms can perpetuate or even amplify those biases. A widely cited study examined a risk assessment algorithm used in criminal sentencing decisions across the United States. The research found that the system was significantly more likely to incorrectly flag Black defendants as high-risk for reoffending compared to white defendants. Because the algorithm was trained on data from a criminal justice system with documented racial disparities, it effectively encoded those disparities into its predictions. The developers of such systems face the difficult task of correcting for historical inequities without introducing new forms of bias through their interventions.

Welfare systems represent another domain where algorithmic governance has expanded rapidly. Several countries now use automated systems to determine benefit eligibility, calculate payment amounts, and flag cases for fraud investigation. The Dutch government implemented the SyRI system (System Risk Indication) to identify potential welfare fraud by analyzing data from multiple government agencies. However, a court ruled in 2020 that the system violated privacy rights and lacked sufficient transparency about its methods. The system was particularly criticized for disproportionately targeting lower-income neighborhoods, raising concerns that algorithmic systems might intensify rather than reduce socioeconomic inequalities.

Despite these challenges, proponents of algorithmic governance identify genuine advantages. Algorithms can process vastly larger datasets than human analysts, potentially identifying patterns and insights that would otherwise remain hidden. They can standardize decisions across thousands of cases, reducing the influence of individual prejudices or inconsistencies. In resource allocation for public services, algorithms can optimize complex systems with multiple variables – determining bus routes, scheduling infrastructure maintenance, or allocating educational resources – with an efficiency that exceeds human capability.

The question, then, is not whether algorithms should play a role in governance but how to ensure that role is properly circumscribed and subject to appropriate oversight. Some scholars advocate for “human-in-the-loop” systems that use AI to analyze data and generate recommendations but reserve final decisions for human officials. Others propose algorithmic impact assessments before deploying AI in policy contexts, similar to environmental impact assessments for development projects. Such assessments would evaluate potential effects on different demographic groups, identify risks of bias, and establish monitoring protocols.

The Australian government’s automated debt recovery system, known colloquially as “Robodebt,” offers a cautionary tale. The system automatically generated debt notices to welfare recipients based on income discrepancies detected by matching tax and welfare records. However, the system’s methodology was fundamentally flawed, assuming that annual income was earned evenly throughout the year – an assumption particularly problematic for seasonal and casual workers. Thousands of citizens received erroneous debt notices, creating significant financial hardship and emotional distress. The government eventually abandoned the system and agreed to repay AUD 721 million in incorrectly raised debts. The failure highlighted how insufficient scrutiny of algorithmic systems before deployment can lead to widespread harm.

Questions 14-18

Choose the correct letter, A, B, C or D.

14. What is the main concern critics have about predictive policing algorithms?

• A) They are too expensive to implement
• B) They may create cycles that reinforce existing patterns
• C) They are less accurate than human predictions
• D) They require too much training data

15. According to the passage, “black box” algorithms are problematic because:

• A) They consume too much energy
• B) They are manufactured by foreign companies
• C) Their decision-making processes cannot be easily understood
• D) They make too many errors

16. The GDPR provision mentioned in the passage:

• A) Bans all automated decision-making
• B) Requires companies to pay fines for using AI
• C) Provides individuals with rights regarding automated decisions
• D) Mandates government approval for all algorithms

17. What problem was identified with the risk assessment algorithm used in criminal sentencing?

• A) It was too expensive to operate
• B) It showed racial bias in its predictions
• C) It incorrectly predicted all defendants
• D) It was too slow to be practical

18. The Dutch SyRI system was discontinued because:

• A) It was too expensive to maintain
• B) It detected too little fraud
• C) Courts found it violated privacy rights
• D) It required too much human oversight

Questions 19-23

Complete the summary below.

Choose NO MORE THAN TWO WORDS from the passage for each answer.

Algorithmic governance involves using computational systems in policy-making, which raises questions about accountability and the balance between human judgment and (19) __. While algorithms can process larger datasets than humans and reduce (20) __ in decision-making, they also face challenges. Many algorithms function as (21) __, making their reasoning difficult to understand. Additionally, when trained on historical data containing discrimination, they can (22) __ existing biases. The Australian Robodebt system demonstrated how (23) __ of algorithmic systems before use can cause widespread harm.

Questions 24-26

Do the following statements agree with the claims of the writer in Passage 2?

Write:

• YES if the statement agrees with the claims of the writer
• NO if the statement contradicts the claims of the writer
• NOT GIVEN if it is impossible to say what the writer thinks about this

24. Algorithms should never be used in government decision-making.
25. Human-in-the-loop systems represent a potential solution for responsible algorithmic governance.
26. All countries have successfully implemented transparent algorithmic systems in welfare administration.

---

PASSAGE 3 – The Geopolitics of Digital Infrastructure and Policy Sovereignty

Độ khó: Hard (Band 7.0-9.0)

Thời gian đề xuất: 23-25 phút

The infrastructural foundations of contemporary governance increasingly rest upon technological architectures that transcend national boundaries, creating novel challenges for policy sovereignty and regulatory authority. As governments worldwide confront the implications of digital transformation, questions about control over data flows, telecommunications networks, and cloud computing platforms have migrated from technical minutiae to central concerns of statecraft and international relations. This shift reflects a broader recognition that the capacity to formulate and implement autonomous policy in domains ranging from national security to economic regulation depends fundamentally upon the technological substrate enabling modern governance.

The concept of “digital sovereignty” has emerged as a framework through which states conceptualize their relationship with transnational technology infrastructure. While definitions vary, the term generally encompasses the ability of a nation to exercise meaningful control over the digital systems and data flows within its jurisdiction, free from external dependencies that might compromise policy autonomy. This concern has intensified as a handful of predominantly American and Chinese technology corporations have achieved unprecedented dominance over critical digital infrastructure. The five largest technology companies – collectively valued at over $8 trillion as of 2021 – wield influence over digital ecosystems that exceeds the regulatory capacity of many nation-states.

Cloud computing infrastructure exemplifies these sovereignty dilemmas. Government agencies increasingly rely on commercial cloud services provided by companies like Amazon Web Services, Microsoft Azure, and Google Cloud for data storage and computational capacity. This outsourcing offers significant advantages in terms of scalability, resilience, and cost-effectiveness compared to maintaining dedicated government data centers. However, it also creates structural dependencies on private corporations, often foreign-owned, for access to data essential to governmental functions. The implications become particularly acute when considering that these services are subject to the legal jurisdiction where the companies are incorporated, potentially requiring them to provide foreign governments access to data they host, even when that data pertains to another nation’s citizens or governmental operations.

Several countries have responded by implementing data localization requirements, mandating that certain categories of data be stored within national borders on locally-owned infrastructure. Russia’s data localization law, enacted in 2015, requires internet companies to store Russian citizens’ personal data on servers physically located within Russia. China’s Cybersecurity Law imposes similar requirements, coupled with extensive provisions for government access to data held by companies operating within its jurisdiction. The European Union has pursued a different approach through instruments like the General Data Protection Regulation, which establishes stringent requirements for data protection while generally permitting cross-border data transfers to jurisdictions deemed to provide “adequate protection.”

Hạ tầng kỹ thuật số và chủ quyền chính sách của các quốc gia trong bối cảnh toàn cầu hóa

These regulatory divergences create fragmentation in what was previously envisioned as a borderless internet, raising concerns about a “splinternet” where different jurisdictions operate fundamentally incompatible digital ecosystems. The economic implications are substantial: compliance costs for companies operating across multiple jurisdictions escalate, economies of scale diminish, and innovation may be impeded by the need to navigate contradictory regulatory requirements. Moreover, authoritarian regimes have exploited data localization requirements to enhance surveillance capabilities and restrict information flows, using sovereignty rhetoric to justify repressive policies.

The telecommunications infrastructure supporting 5G networks has become a particularly contentious arena for these sovereignty concerns. The extensive involvement of Chinese company Huawei in building 5G infrastructure across Europe, Asia, and Africa prompted the United States to launch an aggressive diplomatic campaign urging allies to exclude Huawei equipment from their networks. American officials argued that Huawei’s relationships with the Chinese government created unacceptable risks of espionage or sabotage, potentially compromising the security of communications passing through Huawei equipment. The company and Chinese government vehemently denied these allegations, characterizing them as efforts to preserve American technological dominance by excluding competitive alternatives.

This dispute illuminates how technical infrastructure choices have become enmeshed with geopolitical competition. The decision about which company provides a nation’s telecommunications equipment is not merely a technical or economic calculation but implicates alliance structures, intelligence sharing relationships, and strategic positioning in an increasingly multipolar world. Several countries, including the United Kingdom, Australia, and Japan, partially or entirely restricted Huawei’s participation in their 5G networks, while many developing nations continued utilizing Huawei equipment, attracted by its cost competitiveness and financing arrangements.

Beyond infrastructure ownership, the governance of digital standards represents another dimension where technology influences policy sovereignty. Technical standards – the detailed specifications determining how technologies interoperate – exert profound influence over market structure, competitive dynamics, and the feasibility of regulatory interventions. Historically, international standards bodies like the International Telecommunication Union operated on principles of technical consensus and political neutrality. However, as the strategic significance of digital technologies has become apparent, standards-setting processes have become increasingly politicized, with nations maneuvering to ensure standards reflect their technological capabilities and policy preferences.

The development of facial recognition standards provides an illustrative case. Chinese companies and government representatives have been extensively involved in proposing standards for facial recognition technology through international bodies. Critics argue these standards efforts aim to normalize surveillance practices and create competitive advantages for Chinese facial recognition companies, which have developed sophisticated capabilities partly through deployment in China’s extensive surveillance infrastructure. If international standards reflect Chinese approaches, countries implementing facial recognition systems may find Chinese technology most readily compatible with those standards, creating path dependencies that entrench Chinese technological influence.

The question of how democratic governance can be maintained when essential governmental functions depend on opaque, privately-controlled technological systems represents perhaps the most fundamental challenge posed by technology’s influence on policy. Traditional models of democratic accountability presume that elected officials and public servants make decisions subject to public scrutiny and institutional checks. When critical decision-making increasingly depends on proprietary algorithms, corporate platforms, and infrastructure controlled by entities with potentially conflicting interests, these accountability mechanisms are attenuated. The capacity to formulate genuinely independent policy requires not merely formal legal authority but practical capability – including technical expertise, infrastructural alternatives, and economic resources – that many governments find increasingly difficult to maintain.

Some scholars advocate for a “public option” in digital infrastructure – government-provided or government-supported alternatives to corporate platforms for essential services. The idea draws analogies to traditional public utilities, arguing that just as governments historically ensured access to electricity, water, and telecommunications as public goods, contemporary governance requires ensuring access to digital infrastructure that is not solely subject to corporate control or foreign jurisdiction. Estonia’s X-Road system, a decentralized data exchange platform enabling secure information sharing across government agencies and with the private sector, represents one model. The system, entirely controlled by the Estonian government, has been adopted by several other nations seeking greater technological autonomy.

Questions 27-31

Complete the sentences below.

Choose NO MORE THAN THREE WORDS from the passage for each answer.

27. Digital sovereignty refers to a nation’s ability to exercise control over digital systems without __ that might reduce policy independence.

28. The five largest technology companies have a combined value exceeding __ as of 2021.

29. Russia’s data localization law requires companies to store Russian citizens’ data on __ within Russia.

30. Concerns about a “splinternet” relate to the creation of __ digital ecosystems in different jurisdictions.

31. Estonia’s X-Road system is described as a __ that allows secure information exchange.

Questions 32-36

Do the following statements agree with the information given in Passage 3?

Write:

• TRUE if the statement agrees with the information
• FALSE if the statement contradicts the information
• NOT GIVEN if there is no information on this

32. All government agencies have completely stopped using their own data centers in favor of commercial cloud services.

33. The European Union’s GDPR generally prohibits cross-border data transfers under all circumstances.

34. Huawei denied allegations that it posed security risks to telecommunications networks.

35. International standards bodies have always operated with complete political neutrality.

36. Estonia’s X-Road system has been adopted by every European nation.

Questions 37-40

Choose the correct letter, A, B, C or D.

37. According to the passage, data localization requirements:

• A) Are universally supported by all democratic nations
• B) Have been exploited by authoritarian regimes for surveillance
• C) Have completely solved sovereignty concerns
• D) Are only implemented in developing countries

38. The dispute over Huawei’s involvement in 5G networks demonstrates:

• A) That technical decisions are purely economic
• B) That all countries have identical security concerns
• C) How infrastructure choices relate to geopolitical competition
• D) That telecommunications technology is unimportant

39. The passage suggests that technical standards:

• A) Have no influence on market competition
• B) Are completely neutral and apolitical
• C) Can create competitive advantages for certain technologies
• D) Are irrelevant to policy sovereignty

40. The concept of a “public option” in digital infrastructure:

• A) Is based on analogies to traditional public utilities
• B) Has been universally rejected by all governments
• C) Would eliminate all private technology companies
• D) Is only possible in authoritarian states

---

3. Answer Keys – Đáp Án

PASSAGE 1: Questions 1-13

1. FALSE
2. TRUE
3. TRUE
4. TRUE
5. FALSE
6. NOT GIVEN
7. digital participation
8. two-way communication
9. marginalized communities
10. accountability mechanisms
11. B
12. C
13. C

PASSAGE 2: Questions 14-26

14. B
15. C
16. C
17. B
18. C
19. machine efficiency
20. prejudices / inconsistencies
21. black boxes
22. perpetuate / amplify
23. insufficient scrutiny
24. NO
25. YES
26. NOT GIVEN

PASSAGE 3: Questions 27-40

27. external dependencies
28. $8 trillion
29. servers (physically located)
30. fundamentally incompatible
31. decentralized data exchange platform
32. FALSE
33. FALSE
34. TRUE
35. FALSE
36. NOT GIVEN
37. B
38. C
39. C
40. A

---

4. Giải Thích Đáp Án Chi Tiết

Passage 1 – Giải Thích

Câu 1: FALSE

• Dạng câu hỏi: True/False/Not Given
• Từ khóa: Traditional consultation methods, completely abandoned, digital platforms
• Vị trí trong bài: Đoạn 1, dòng 2-4
• Giải thích: Câu hỏi nói “completely being abandoned” (hoàn toàn bị từ bỏ), nhưng bài đọc chỉ nói “are increasingly being supplemented – and sometimes replaced” (ngày càng được bổ sung – và đôi khi thay thế). Từ “supplemented” chứng tỏ phương pháp truyền thống vẫn còn được sử dụng, không phải hoàn toàn bị loại bỏ. Đây là ví dụ điển hình của paraphrase với từ cực đoan “completely” làm thay đổi nghĩa.

Câu 2: TRUE

• Dạng câu hỏi: True/False/Not Given
• Từ khóa: Decide Madrid platform, proposals, citizen support, public vote
• Vị trí trong bài: Đoạn 2, dòng 5-7
• Giải thích: Bài đọc nói rõ: “If a proposal receives enough support from other citizens, it advances to a public vote”. Đây là paraphrase trực tiếp với “enough support” = “sufficient citizen support” và “advances to” = “proceeding to”.

Câu 3: TRUE

• Dạng câu hỏi: True/False/Not Given
• Từ khóa: Social media, effective, traditional media, earthquake warnings, Japan 2011
• Vị trí trong bài: Đoạn 3, dòng 5-7
• Giải thích: Bài viết nói: “automated alerts reaching citizens faster than traditional broadcast media”. “Faster” chứng tỏ hiệu quả hơn trong việc cung cấp cảnh báo sớm.

Câu 4: TRUE

• Dạng câu hỏi: True/False/Not Given
• Từ khóa: Digital divide, demographic groups
• Vị trí trong bài: Đoạn 4, dòng 4-8
• Giải thích: Bài đọc liệt kê cụ thể: “Elderly populations, rural residents with poor connectivity, and low-income households” – đây là các nhóm nhân khẩu học (demographic groups) bị ảnh hưởng nhiều hơn.

Câu 5: FALSE

• Dạng câu hỏi: True/False/Not Given
• Từ khóa: Professional facilitators, only method, moderate online discussions
• Vị trí trong bài: Đoạn 5, dòng 5-7
• Giải thích: Bài đọc nói “Some jurisdictions employ professional facilitators… while others rely on community moderation models”. Từ “while others” chứng tỏ có nhiều phương pháp, không chỉ một.

Câu 7: digital participation

• Dạng câu hỏi: Sentence Completion
• Từ khóa: E-democracy, various forms
• Vị trí trong bài: Đoạn 2, câu đầu
• Giải thích: Câu gốc: “E-democracy… encompasses various forms of digital participation in political processes”. Câu hỏi paraphrase “includes” thành “encompasses”.

Câu 11: B

• Dạng câu hỏi: Multiple Choice
• Từ khóa: Data analytics, benefit
• Vị trí trong bài: Đoạn 6
• Giải thích: Bài viết nói: “Governments can now analyze patterns in citizen feedback at scales previously impossible” và “AI tools can process thousands of survey responses, identifying common themes”. Đáp án B “identify patterns in large volumes of citizen feedback” là paraphrase chính xác nhất.

Passage 2 – Giải Thích

Câu 14: B

• Dạng câu hỏi: Multiple Choice
• Từ khóa: Critics, predictive policing algorithms, concern
• Vị trí trong bài: Đoạn 2, dòng 6-10
• Giải thích: Bài viết nói về “self-fulfilling prophecies” và “cycles of over-policing”. Đáp án B “create cycles that reinforce existing patterns” paraphrase chính xác ý này.

Câu 17: B

• Dạng câu hỏi: Multiple Choice
• Từ khóa: Risk assessment algorithm, criminal sentencing, problem
• Vị trí trong bài: Đoạn 5, dòng 3-7
• Giải thích: Bài viết nói rõ: “significantly more likely to incorrectly flag Black defendants as high-risk for reoffending compared to white defendants”. Đây là racial bias (thiên kiến chủng tộc).

Câu 19: machine efficiency

• Dạng câu hỏi: Summary Completion
• Từ khóa: Balance between human judgment and…
• Vị trí trong bài: Đoạn 1, câu cuối
• Giải thích: Câu gốc: “the proper balance between human judgment and machine efficiency”.

Câu 21: black boxes

• Dạng câu hỏi: Summary Completion
• Từ khóa: Algorithms function as…
• Vị trí trong bài: Đoạn 3, dòng 2-3
• Giải thích: Bài viết sử dụng thuật ngữ “black boxes” để mô tả các thuật toán không thể hiểu được.

Câu 25: YES

• Dạng câu hỏi: Yes/No/Not Given
• Từ khóa: Human-in-the-loop systems, potential solution
• Vị trí trong bài: Đoạn 8
• Giải thích: Tác giả viết: “Some scholars advocate for ‘human-in-the-loop’ systems” và giải thích rõ cách hoạt động. Việc đề cập trong phần thảo luận giải pháp cho thấy tác giả đồng ý đây là một giải pháp tiềm năng.

Passage 3 – Giải Thích

Câu 27: external dependencies

• Dạng câu hỏi: Sentence Completion
• Từ khóa: Digital sovereignty, control, without…
• Vị trí trong bài: Đoạn 2, dòng 2-4
• Giải thích: Câu gốc: “free from external dependencies that might compromise policy autonomy”.

Câu 32: FALSE

• Dạng câu hỏi: True/False/Not Given
• Từ khóa: All government agencies, completely stopped, own data centers
• Vị trí trong bài: Đoạn 3, dòng 1-2
• Giải thích: Bài viết nói “Government agencies increasingly rely on commercial cloud services” (ngày càng dựa vào), chứ không nói “all” (tất cả) hay “completely stopped” (hoàn toàn ngừng). Từ “increasingly” chỉ xu hướng, không phải hiện trạng tuyệt đối.

Câu 34: TRUE

• Dạng câu hỏi: True/False/Not Given
• Từ khóa: Huawei, denied allegations, security risks
• Vị trí trong bài: Đoạn 6, dòng 5-6
• Giải thích: Bài viết nói rõ: “The company and Chinese government vehemently denied these allegations”.

Câu 38: C

• Dạng câu hỏi: Multiple Choice
• Từ khóa: Dispute over Huawei, 5G networks, demonstrates
• Vị trí trong bài: Đoạn 7
• Giải thích: Câu chủ đề của đoạn 7: “This dispute illuminates how technical infrastructure choices have become enmeshed with geopolitical competition”. Đáp án C paraphrase chính xác ý này.

Câu 40: A

• Dạng câu hỏi: Multiple Choice
• Từ khóa: Public option, digital infrastructure, concept based on
• Vị trí trong bài: Đoạn 11, dòng 1-4
• Giải thích: Bài viết nói: “The idea draws analogies to traditional public utilities”. Đáp án A “based on analogies to traditional public utilities” là paraphrase trực tiếp.

---

5. Từ Vựng Quan Trọng Theo Passage

Passage 1 – Essential Vocabulary

Từ vựng	Loại từ	Phiên âm	Nghĩa tiếng Việt	Ví dụ từ bài	Collocation
dramatic transformation	n. phrase	/drəˈmætɪk ˌtrænsfərˈmeɪʃən/	sự chuyển đổi mạnh mẽ	“has undergone a dramatic transformation”	undergo/experience a dramatic transformation
public consultation	n. phrase	/ˈpʌblɪk ˌkɒnsəlˈteɪʃən/	tham vấn công chúng	“Traditional methods of public consultation”	conduct/hold public consultation
digital platforms	n. phrase	/ˈdɪdʒɪtəl ˈplætfɔːmz/	nền tảng kỹ thuật số	“supplemented by digital platforms”	develop/launch digital platforms
policy formulation	n. phrase	/ˈpɒləsi ˌfɔːmjuˈleɪʃən/	xây dựng chính sách	“contribute directly to policy formulation”	participate in policy formulation
direct democracy	n. phrase	/dɪˈrekt dɪˈmɒkrəsi/	dân chủ trực tiếp	“This direct democracy approach”	promote/practice direct democracy
two-way communication	n. phrase	/tuː-weɪ kəˌmjuːnɪˈkeɪʃən/	giao tiếp hai chiều	“engage in two-way communication”	establish/facilitate two-way communication
crowd-sourced information	n. phrase	/kraʊd-sɔːst ˌɪnfəˈmeɪʃən/	thông tin từ đám đông	“gather crowd-sourced information”	collect/utilize crowd-sourced information
digital divide	n. phrase	/ˈdɪdʒɪtəl dɪˈvaɪd/	khoảng cách kỹ thuật số	“The digital divide”	bridge/narrow the digital divide
marginalized communities	n. phrase	/ˈmɑːdʒɪnəlaɪzd kəˈmjuːnətiz/	cộng đồng bị thiệt thòi	“may inadvertently exclude marginalized communities”	support/empower marginalized communities
data analytics	n. phrase	/ˈdeɪtə ˌænəˈlɪtɪks/	phân tích dữ liệu	“Data analytics represents another dimension”	apply/use data analytics
accountability mechanisms	n. phrase	/əˌkaʊntəˈbɪləti ˈmekənɪzəmz/	cơ chế trách nhiệm	“creates accountability mechanisms”	establish/strengthen accountability mechanisms
responsible stewardship	n. phrase	/rɪˈspɒnsəbl ˈstjuːədʃɪp/	quản lý có trách nhiệm	“demonstrate responsible stewardship”	exercise responsible stewardship

Passage 2 – Essential Vocabulary

Từ vựng	Loại từ	Phiên âm	Nghĩa tiếng Việt	Ví dụ từ bài	Collocation
algorithmic governance	n. phrase	/ˌælɡəˈrɪðmɪk ˈɡʌvənəns/	quản trị thuật toán	“algorithmic governance involves using computational systems”	implement/study algorithmic governance
decision-making processes	n. phrase	/dɪˈsɪʒən-ˌmeɪkɪŋ ˈprəʊsesɪz/	quy trình ra quyết định	“into governmental decision-making processes”	improve/streamline decision-making processes
predictive policing	n. phrase	/prɪˈdɪktɪv pəˈliːsɪŋ/	cảnh sát dự đoán	“Predictive policing provides one example”	deploy/implement predictive policing
self-fulfilling prophecy	n. phrase	/ˌself fʊlˈfɪlɪŋ ˈprɒfəsi/	lời tiên tri tự thực hiện	“concerns about self-fulfilling prophecies”	create/become a self-fulfilling prophecy
black box	n. phrase	/blæk bɒks/	hộp đen (không thể hiểu)	“function as ‘black boxes'”	operate as/remain a black box
algorithmic accountability	n. phrase	/ˌælɡəˈrɪðmɪk əˌkaʊntəˈbɪləti/	trách nhiệm thuật toán	“pioneering approaches to algorithmic accountability”	ensure/promote algorithmic accountability
automated decision-making	n. phrase	/ˈɔːtəmeɪtɪd dɪˈsɪʒən-ˌmeɪkɪŋ/	ra quyết định tự động	“subject to automated decision-making”	rely on/regulate automated decision-making
algorithmic bias	n. phrase	/ˌælɡəˈrɪðmɪk ˈbaɪəs/	thiên kiến thuật toán	“Algorithmic bias presents another challenge”	detect/eliminate algorithmic bias
historical data	n. phrase	/hɪˈstɒrɪkəl ˈdeɪtə/	dữ liệu lịch sử	“AI systems learn from historical data”	analyze/train on historical data
perpetuate	v.	/pəˈpetʃueɪt/	duy trì, làm trường tồn	“algorithms can perpetuate those biases”	perpetuate stereotypes/discrimination
risk assessment	n. phrase	/rɪsk əˈsesmənt/	đánh giá rủi ro	“a risk assessment algorithm”	conduct/perform risk assessment
racial disparities	n. phrase	/ˈreɪʃəl dɪˈspærətiz/	bất bình đẳng chủng tộc	“documented racial disparities”	address/reduce racial disparities
benefit eligibility	n. phrase	/ˈbenɪfɪt ˌelɪdʒəˈbɪləti/	điều kiện hưởng trợ cấp	“determine benefit eligibility”	assess/verify benefit eligibility
fraud investigation	n. phrase	/frɔːd ɪnˌvestɪˈɡeɪʃən/	điều tra gian lận	“flag cases for fraud investigation”	conduct/initiate fraud investigation
human-in-the-loop	adj. phrase	/ˈhjuːmən-ɪn-ðə-luːp/	có sự tham gia của con người	“human-in-the-loop systems”	implement/design human-in-the-loop systems

Passage 3 – Essential Vocabulary

Từ vựng	Loại từ	Phiên âm	Nghĩa tiếng Việt	Ví dụ từ bài	Collocation
infrastructural foundations	n. phrase	/ˌɪnfrəˈstrʌktʃərəl faʊnˈdeɪʃənz/	nền tảng hạ tầng	“The infrastructural foundations of contemporary governance”	establish/build infrastructural foundations
technological architectures	n. phrase	/ˌteknəˈlɒdʒɪkəl ˈɑːkɪtektʃəz/	kiến trúc công nghệ	“rest upon technological architectures”	design/develop technological architectures
policy sovereignty	n. phrase	/ˈpɒləsi ˈsɒvrənti/	chủ quyền chính sách	“challenges for policy sovereignty”	maintain/protect policy sovereignty
regulatory authority	n. phrase	/ˈreɡjələtəri ɔːˈθɒrəti/	thẩm quyền quản lý	“policy sovereignty and regulatory authority”	exercise/establish regulatory authority
digital transformation	n. phrase	/ˈdɪdʒɪtəl ˌtrænsfəˈmeɪʃən/	chuyển đổi số	“confront the implications of digital transformation”	undergo/drive digital transformation
statecraft	n.	/ˈsteɪtkrɑːft/	nghệ thuật quản trị nhà nước	“central concerns of statecraft”	exercise/practice statecraft
digital sovereignty	n. phrase	/ˈdɪdʒɪtəl ˈsɒvrənti/	chủ quyền số	“The concept of ‘digital sovereignty'”	assert/defend digital sovereignty
transnational	adj.	/ˌtrænzˈnæʃənəl/	xuyên quốc gia	“relationship with transnational technology infrastructure”	transnational corporations/networks
external dependencies	n. phrase	/ɪkˈstɜːnəl dɪˈpendənsiz/	sự phụ thuộc bên ngoài	“free from external dependencies”	reduce/eliminate external dependencies
unprecedented dominance	n. phrase	/ʌnˈpresɪdentɪd ˈdɒmɪnəns/	sự thống trị chưa từng có	“achieved unprecedented dominance”	establish/maintain unprecedented dominance
cloud computing	n. phrase	/klaʊd kəmˈpjuːtɪŋ/	điện toán đám mây	“Cloud computing infrastructure”	adopt/utilize cloud computing
scalability	n.	/ˌskeɪləˈbɪləti/	khả năng mở rộng	“advantages in terms of scalability”	ensure/improve scalability
data localization	n. phrase	/ˈdeɪtə ˌləʊkəlaɪˈzeɪʃən/	địa phương hóa dữ liệu	“implementing data localization requirements”	mandate/enforce data localization
cross-border data transfers	n. phrase	/krɒs-ˈbɔːdə ˈdeɪtə ˈtrænsfɜːz/	chuyển dữ liệu xuyên biên giới	“generally permitting cross-border data transfers”	regulate/restrict cross-border data transfers
fragmentation	n.	/ˌfræɡmenˈteɪʃən/	sự phân mảnh	“create fragmentation”	cause/prevent fragmentation
splinternet	n.	/ˈsplɪntənet/	internet bị phân mảnh	“concerns about a ‘splinternet'”	result in/lead to splinternet
telecommunications infrastructure	n. phrase	/ˌtelikəˌmjuːnɪˈkeɪʃənz ˌɪnfrəˈstrʌktʃə/	hạ tầng viễn thông	“The telecommunications infrastructure supporting 5G”	build/upgrade telecommunications infrastructure
geopolitical competition	n. phrase	/ˌdʒiːəʊpəˈlɪtɪkəl ˌkɒmpəˈtɪʃən/	cạnh tranh địa chính trị	“enmeshed with geopolitical competition”	intensify/engage in geopolitical competition
technical standards	n. phrase	/ˈteknɪkəl ˈstændədz/	tiêu chuẩn kỹ thuật	“governance of digital standards”	develop/establish technical standards
interoperate	v.	/ˌɪntərˈɒpəreɪt/	tương tác, hoạt động cùng nhau	“how technologies interoperate”	systems interoperate/can interoperate
path dependencies	n. phrase	/pɑːθ dɪˈpendənsiz/	sự phụ thuộc vào lộ trình	“creating path dependencies”	create/establish path dependencies
democratic accountability	n. phrase	/ˌdeməˈkrætɪk əˌkaʊntəˈbɪləti/	trách nhiệm dân chủ	“Traditional models of democratic accountability”	ensure/maintain democratic accountability
proprietary algorithms	n. phrase	/prəˈpraɪətəri ˈælɡərɪðəmz/	thuật toán độc quyền	“depends on proprietary algorithms”	develop/protect proprietary algorithms
public utilities	n. phrase	/ˈpʌblɪk juːˈtɪlətiz/	dịch vụ công cộng	“analogies to traditional public utilities”	provide/manage public utilities

---

Kết bài

Chủ đề “How is technology influencing public policy?” không chỉ là một đề tài thú vị trong học thuật mà còn phản ánh thực tế đang diễn ra khắp thế giới. Qua bộ đề thi IELTS Reading này, bạn đã được trải nghiệm đầy đủ 3 passages với độ khó tăng dần, từ việc tìm hiểu về nền tảng dân chủ kỹ thuật số ở cấp độ cơ bản, đến quản trị thuật toán ở mức trung bình, và cuối cùng là địa chính trị hạ tầng số ở mức độ nâng cao.

Đề thi này đã cung cấp cho bạn 40 câu hỏi đa dạng với 7 dạng bài khác nhau, giúp bạn làm quen với mọi format có thể xuất hiện trong IELTS Reading thực tế. Phần đáp án chi tiết kèm giải thích đã chỉ ra cách paraphrase giữa câu hỏi và passage, vị trí chính xác của thông tin, và lý do tại sao các đáp án khác không phù hợp – những kỹ năng thiết yếu để đạt band điểm cao.

Bộ từ vựng được tổng hợp từ 3 passages không chỉ giúp bạn hiểu sâu hơn về chủ đề mà còn cung cấp collocations và cách sử dụng trong ngữ cảnh học thuật. Hãy học thuộc những từ này vì chúng có thể xuất hiện trong cả phần Writing Task 2 và Speaking Part 3 khi bạn thảo luận về công nghệ và xã hội.

Hãy nhớ rằng, việc làm đề thi mẫu chỉ có giá trị khi bạn tự phân tích lỗi sai, hiểu tại sao mình chọn sai, và rút ra bài học cho lần sau. Đối với IELTS Reading, kỹ thuật làm bài quan trọng không kém gì vốn từ vựng. Chúc bạn ôn tập hiệu quả và đạt band điểm mục tiêu!