디지털 혁신 영어

파트1


1. A company updates a wearable device after launch so that it can detect a new type of health signal without changing the hardware. Which concept does this most directly illustrate?
  • 1
     Reprogrammability
  • 2
     Generativity
  • 3
     Datafication
  • 4
     Scalability

2. A platform's users begin using its tools in unexpected ways and build new add-on services that the original company never planned. Which concept does this best illustrate?
  • 1
     Reprogrammability
  • 2
     Generativity
  • 3
     Datafication
  • 4
     Scalability

3. A firm says, "Our biggest strength is that every user interaction leaves a digital trace that we can analyze for insight and control." Which concept is the firm emphasizing?
  • 1
     Reprogrammability
  • 2
     Generativity
  • 3
     Datafication
  • 4
     Scalability

4. A digital product reaches millions of users with almost no additional distribution cost, but the same process also spreads bugs and bias very quickly. Which concept is most directly involved?
  • 1
     Reprogrammability
  • 2
     Generativity
  • 3
     Datafication
  • 4
     Scalability

5. Which statement best captures the difference between reprogrammability and generativity?
  • 1
     Reprogrammability comes from outside users, while generativity comes from software updates by the firm.
  • 2
     Reprogrammability and generativity both mean that digital products can be copied cheaply.
  • 3
     Reprogrammability is about a product's functions changing after launch, while generativity is about outsiders creating new value beyond the firm's original intentions.
  • 4
     Reprogrammability is about data collection, while generativity is about organizational alignment.

6. Which statement best captures the difference between datafication and scalability?
  • 1
     Datafication is about low-cost replication, while scalability is about turning usage into data.
  • 2
     Datafication turns activities into digital data, while scalability allows digital outputs to spread widely at near-zero marginal cost.
  • 3
     Datafication depends on outsiders, while scalability depends on managers.
  • 4
     Datafication and scalability both refer to the same growth mechanism.

7. A manager says, "Our digital transformation failed because the software was not advanced enough." Based on the study guide's logic, what is the strongest challenge to this claim?
  • 1
     Digital innovation fails only when products cannot scale globally.
  • 2
     Digital innovation is mainly about product design, not about work or incentives.
  • 3
     Even if the technology works, innovation can still fail when management, strategy, and organizational alignment are weak.
  • 4
     If outside users are not involved, digital innovation is impossible.

8. Which statement best reflects why digital innovation is called socio-technical?
  • 1
     Because social media is always required for digital products to succeed.
  • 2
     Because technical performance matters, but organizational structures, incentives, and labor practices matter too.
  • 3
     Because digital innovation replaces management with automation.
  • 4
     Because every digital product must be built by an external ecosystem.

파트2


9. A platform reduces opportunistic behavior by using ratings, escrow, and standardized digital contracts. Which ACE function is this most directly addressing?
  • 1
     Alignment of Incentives
  • 2
     Coordination of Activities
  • 3
     Expertise Integration
  • 4
     Scalability

10. A company uses algorithmic matching, synchronized calendars, and modular task structures to make distributed work run smoothly. Which ACE function is most directly involved?
  • 1
     Alignment of Incentives
  • 2
     Coordination of Activities
  • 3
     Expertise Integration
  • 4
     Generativity

11. A digital firm relies on online communities, shared workspaces, and APIs so that different kinds of knowledge can be combined for learning and innovation. Which ACE function does this best illustrate?
  • 1
     Alignment of Incentives
  • 2
     Coordination of Activities
  • 3
     Expertise Integration
  • 4
     Resource Control

12. Which statement best captures the idea of the "inverted firm"?
  • 1
     Firms create more value by moving activities back inside the organization and expanding managerial control.
  • 2
     Digitalization makes markets less efficient, so firms increasingly replace external networks with hierarchy.
  • 3
     Value-creating activities move outside firm boundaries, while the firm increasingly orchestrates external networks.
  • 4
     Firms become inverted when managers are removed and no coordination function is needed.

13. According to the study guide, why did traditional hierarchical firms historically outperform markets in many settings?
  • 1
     Markets lacked consumers, so firms had no competition.
  • 2
     Managers were costless and therefore always more efficient than digital systems.
  • 3
     Hierarchies could perform alignment, coordination, and expertise integration more effectively through managerial oversight and shared culture.
  • 4
     Firms were better only because they owned more physical assets.

14. Which statement best explains why digitalization can make market contracting more attractive?
  • 1
     It eliminates the need for coordination and expertise integration.
  • 2
     It makes managers more powerful and lowers the value of reputation systems.
  • 3
     It guarantees that external markets will always outperform firms.
  • 4
     It allows markets to perform ACE functions more effectively through tools like reputation systems, digital workspaces, and modular architectures.

15. A student says, "Alignment of incentives and coordination of activities are basically the same thing." Which response best fits the study guide?
  • 1
     Correct, because both are mainly about matching workers to tasks.
  • 2
     Correct, because both are forms of expertise integration.
  • 3
     Incorrect, because alignment reduces opportunism, while coordination manages information and interdependence among activities.
  • 4
     Incorrect, because alignment is only relevant inside firms, while coordination exists only in markets.

16. Which comparison between hierarchy and digital market mechanisms is correctly matched?
  • 1
     Hierarchy - online communities and shared digital workspaces for expertise integration
  • 2
     Digital market - CEO intervention and direct managerial oversight for incentive alignment
  • 3
     Hierarchy - algorithmic matching and modular task structures for coordination
  • 4
     Digital market - reputation systems and escrow accounts for alignment of incentives

파트3


17. A student says, "A platform is just a pipeline with more users." Which response best fits the study guide?
  • 1
     Correct, because both models mainly rely on internal resource control and linear value delivery.
  • 2
     Correct, because the only real difference is the size of the customer base.
  • 3
     Incorrect, because a pipeline produces and sells through a linear chain, while a platform orchestrates interactions among external participants.
  • 4
     Incorrect, because platforms do not create value and only capture value from producers.

18. Which statement best captures the shift from resource control to resource orchestration?
  • 1
     The firm increases ownership of all critical assets so it can eliminate outside dependence.
  • 2
     The firm gives up governance and lets producers and consumers organize themselves.
  • 3
     The firm focuses less on owning resources directly and more on coordinating outside participants who create ecosystem value.
  • 4
     The firm shifts from customer value to cost minimization.

19. Which example best illustrates modularity?
  • 1
     A system works only when every component is redesigned together as one tightly integrated unit.
  • 2
     A component's output can be defined and connected through standard interfaces without knowing other components' internal processes.
  • 3
     A platform charges all participants the same fee to simplify pricing.
  • 4
     A firm vertically integrates production to reduce transaction costs.

20. Which situation fails the study guide's test for independence?
  • 1
     A developer can specify a module's output without referring to another module's internal logic.
  • 2
     A service can be plugged into the system through an API with limited dependence on other components.
  • 3
     A component can only be defined by referring to the inputs and internal workings of several other components.
  • 4
     A feature can be added without redesigning the rest of the system.

21. A platform leader first expands ecosystem participation and customer satisfaction, then secures an unavoidable control point to capture value. What sequence is the study guide describing?
  • 1
     Cost reduction first, then internal optimization
  • 2
     Value capture first, then value creation
  • 3
     Value creation first, then value capture through bottlenecks
  • 4
     Profit maximization first, then network effects

22. Which of the following is the best example of a strategic bottleneck?
  • 1
     A part of the ecosystem that participants can easily bypass or replace
  • 2
     A feature that helps user growth but gives the firm no control over ecosystem interactions
  • 3
     A scarce and unavoidable control point that makes the orchestrator difficult to replace
  • 4
     A low-margin service that many competitors can offer identically

23. In the Android example, which mapping is correct?
  • 1
     Owner - handset maker / Provider - Google / Producer - user / Consumer - app developer
  • 2
     Owner - app developer / Provider - Google Play / Producer - handset maker / Consumer - advertiser
  • 3
     Owner - Google / Provider - handset device / Producer - app developer / Consumer - app user
  • 4
     Owner - Google / Provider - app developer / Producer - handset device / Consumer - operating system

24. A student argues that because Android is more open than iOS, Google does not really control the ecosystem. Which response is most consistent with the study guide?
  • 1
     Correct, because openness means the owner gives up all major control points.
  • 2
     Correct, because bottlenecks can only exist in closed ecosystems like Apple.
  • 3
     Incorrect, because Google still controls key bottlenecks such as OS ownership, the Play Store, proprietary services, and data/search integration.
  • 4
     Incorrect, but only because Google fully controls handset manufacturing.

25. Which statement best distinguishes Apple and Google in the study guide?
  • 1
     Apple emphasizes market share through open horizontal integration, while Google emphasizes profitability through controlled vertical integration.
  • 2
     Apple and Google mainly differ in pricing, not in ecosystem structure.
  • 3
     Apple uses a more controlled, vertically integrated model for profitability, while Google uses a more open, horizontally integrated model for market share.
  • 4
     Google avoids bottlenecks, while Apple depends entirely on hardware margins.

파트4


26. A student says, "Machine Learning is basically the same thing as Artificial Intelligence." Which response best matches the study guide?
  • 1
     Correct, because both terms refer only to neural networks trained on large datasets.
  • 2
     Correct, because AI is just the newest name for machine learning.
  • 3
     Incorrect, because AI is the broader field and ML is a subset focused on learning patterns from data.
  • 4
     Incorrect, because ML is broader than AI and includes all intelligent systems.

27. Which statement best distinguishes deep learning from machine learning in the study guide?
  • 1
     Deep learning refers to any kind of human judgment used to label training data.
  • 2
     Deep learning is a non-learning system that relies on hard-coded rules.
  • 3
     Deep learning is a type of machine learning that uses multi-layered neural networks.
  • 4
     Deep learning is another term for foundation models only.

28. A model is pretrained on massive broad data and later adapted to many downstream tasks. According to the study guide, this is best described as:
  • 1
     a traditional expert system
  • 2
     a digital platform bottleneck
  • 3
     a foundation model
  • 4
     a pure prediction market

29. Which statement best captures the relationship between foundation models and LLMs?
  • 1
     Foundation models are a subtype of LLMs used only for text generation.
  • 2
     LLMs and foundation models are unrelated categories.
  • 3
     LLMs are one subtype of foundation models.
  • 4
     Foundation models are older systems that exclude language tasks.

30. A student says, "Emergence means one model architecture can be used across many domains." What is the best correction?
  • 1
     Correct, because emergence and homogenization mean the same thing.
  • 2
     Partly correct, because emergence refers to standardized training data.
  • 3
     Incorrect, because that is homogenization; emergence refers to capabilities appearing that were not explicitly coded.
  • 4
     Incorrect, because emergence refers only to lower hallucination rates.

31. Which example best illustrates emergence?
  • 1
     A single model architecture is reused for law, medicine, coding, and education.
  • 2
     A company collects cleaner labeled data for model evaluation.
  • 3
     A model displays useful capabilities that were not directly programmed into it.
  • 4
     A hospital replaces human labels with objective physical measurements.

32. Which statement best reflects the study guide's view of predictive AI versus generative AI?
  • 1
     The difference is always determined only by model architecture.
  • 2
     Predictive AI uses data, while generative AI does not.
  • 3
     The distinction can depend on how humans use the system, not just on the model itself.
  • 4
     Generative AI is always foundation-model based, while predictive AI never is.

33. Why is ground truth so central to predictive AI?
  • 1
     Because it replaces the need for evaluation once a model is deployed.
  • 2
     Because larger models no longer need labels if they have enough parameters.
  • 3
     Because it provides the benchmark labels used to train and validate model performance.
  • 4
     Because it guarantees that human bias is removed from all datasets.

34. In medicine, why can using expert opinion as a substitute for ground truth be risky?
  • 1
     Because expert opinion is always random and unusable.
  • 2
     Because expert opinion makes models too creative.
  • 3
     Because the model may learn to reproduce human limitations or bias rather than an ideal standard of judgment.
  • 4
     Because expert opinion automatically prevents hallucinations.

35. Which managerial question best reflects the study guide's advice to "peel back the layers" of AI performance claims?
  • 1
     "How many users does the product have right now?"
  • 2
     "Is the interface more attractive than competing tools?"
  • 3
     "What exactly is being used as ground truth, and does it match the ideal decision standard?"
  • 4
     "Can the model be deployed without any human oversight?"

파트5


36. A task has high-quality ground truth, the underlying process is predictable, and the cost of error is low. According to the study guide, what is the strongest recommendation?
  • 1
     Full automation with predictive AI is often appropriate.
  • 2
     Avoid using AI because humans always outperform machines in low-error settings.
  • 3
     Use augmentation only, because full automation is never recommended.
  • 4
     Use generative AI only, because predictive AI is too rigid.

37. Which statement best reflects the study guide's view of augmentation?
  • 1
     Augmentation always solves the weaknesses of automation.
  • 2
     Augmentation means removing human judgment from the final decision.
  • 3
     Augmentation can help with nuance and accountability, but it may also raise decision costs and fail when humans resist or override AI poorly.
  • 4
     Augmentation is only useful when ground truth is perfect.

38. A manager says, "If the cost of error is high, we should still automate because AI is more consistent than people." Which response best fits the study guide?
  • 1
     Correct, because consistency matters more than the stakes of the task.
  • 2
     Correct, because high error cost mainly matters for generative AI, not predictive AI.
  • 3
     Incorrect, because high error cost is a reason to be cautious about full automation, especially when judgment and accountability are important.
  • 4
     Incorrect, but only because AI should never be used in important tasks.

39. Which task best fits the "No Regrets" zone in the Anand & Wu framework?
  • 1
     Setting corporate strategy during a crisis
  • 2
     Making disciplinary decisions about employees
  • 3
     Screening resumes or summarizing standard documents
  • 4
     Negotiating a highly customized acquisition deal

40. Which task best fits the "Creative Catalyst" zone?
  • 1
     Writing a first draft of ad copy or outlining a sales script
  • 2
     Final approval of a high-value legal contract
  • 3
     Diagnosing rare cancers using uncertain labels
  • 4
     Making layoffs during a reorganization

41. Which task best fits the "Quality Control" zone?
  • 1
     Drafting production code or a high-value contract where mistakes are costly
  • 2
     Brainstorming campaign slogans for a low-risk test
  • 3
     Summarizing meeting notes for internal convenience
  • 4
     Choosing a movie recommendation for a user

42. Which statement best explains why AI productivity gains may fail to improve the bottom line?
  • 1
     Because AI can improve output quality only in tacit-knowledge tasks
  • 2
     Because productivity gains automatically disappear when adoption grows
  • 3
     Because firms may save labor time but fail to redesign workflows or redeploy that capacity into higher-value work
  • 4
     Because bottom-line gains come only from firing workers immediately

43. Which statement best reflects the study guide's view of GenAI governance?
  • 1
     Governance is optional as long as employees find the tools useful.
  • 2
     Governance mainly matters for future regulation, not for current strategy.
  • 3
     Governance is a strategic priority, but it is hard because GenAI use is widespread and difficult to audit, especially on personal devices.
  • 4
     Governance is easy if a company bans all public AI tools.

44. Which comparison between Centaurs, Cyborgs, and Self-Automators is correct?
  • 1
     Centaurs give both the what and how to AI, while Self-Automators selectively use AI for sub-tasks.
  • 2
     Cyborgs avoid iterative collaboration with AI, while Centaurs rely on deep back-and-forth.
  • 3
     Self-Automators develop the strongest domain expertise because they maximize efficiency.
  • 4
     Centaurs keep control of both what and how while using AI selectively, Cyborgs shape the how through iterative collaboration, and Self-Automators hand over both what and how to AI.

파트6


45. A buyer wants a platform that mainly connects them with freelancers, while the buyer still evaluates the final deliverable and manages the project. Which Human Cloud model best fits this case?
  • 1
     Facilitator
  • 2
     Arbitrator
  • 3
     Aggregator
  • 4
     Governor

46. A platform runs a contest in which many suppliers submit solutions, and the buyer selects the best one. Which model is this?
  • 1
     Facilitator
  • 2
     Arbitrator
  • 3
     Aggregator
  • 4
     Governor

47. A platform collects many small standardized inputs and combines them into one output using rules or filters specified by the buyer. Which model best fits this case?
  • 1
     Facilitator
  • 2
     Arbitrator
  • 3
     Aggregator
  • 4
     Governor

48. A platform certifies supplier quality and governs the project so that buyers can trust the platform rather than each individual supplier. Which model is this?
  • 1
     Facilitator
  • 2
     Arbitrator
  • 3
     Aggregator
  • 4
     Governor

49. Which statement best distinguishes Facilitator from Governor?
  • 1
     Facilitator and Governor both leave all quality control to the buyer.
  • 2
     Facilitator is mainly for microtasks, while Governor is only for contests.
  • 3
     Facilitator mainly connects market participants, while Governor adds stronger platform control over project execution and quality assurance.
  • 4
     Governor removes the need for any supplier evaluation or platform operations.

50. Why is circumvention such an important challenge for Facilitator platforms?
  • 1
     Because suppliers in facilitator models are usually banned from talking to buyers.
  • 2
     Because the platform owns all project outputs and must transfer them manually.
  • 3
     Because buyers and freelancers may move their relationship off-platform to avoid platform fees once they find each other.
  • 4
     Because the platform guarantees all project outcomes and becomes legally responsible for every error.

51. Which condition most strongly supports moving from a Facilitator model toward a Governor model?
  • 1
     Tasks are highly ambiguous, tacit, and impossible to specify.
  • 2
     Outcomes are poorly defined, making algorithmic oversight difficult.
  • 3
     Work outcomes are well specified and tasks are modular enough for algorithmic management to coordinate execution more tightly.
  • 4
     Buyers want less service and more freedom to manage everything themselves.

52. What is the best reason larger clients may push a platform toward managed services?
  • 1
     Larger clients usually prefer to inspect thousands of freelancers individually.
  • 2
     Larger clients want lower trust in the platform and more trust in unknown suppliers.
  • 3
     Larger clients often want more reliable quality, coordination, and service guarantees than a pure facilitator model can easily provide.
  • 4
     Larger clients mainly prefer competition models like Kaggle for every kind of work.

53. According to the study guide, how can AI amplify Human Cloud model choices?
  • 1
     AI makes all four models collapse into a single universal structure.
  • 2
     AI matters only for pricing, not for governance or task design.
  • 3
     Predictive AI can improve matching precision, and GenAI can improve task specification, making some forms of tighter platform governance more feasible.
  • 4
     AI eliminates the need for buyers, because the platform can always complete the work itself.

파트7


54. A group solves a problem better than any one member could alone. According to the study guide, this is best described as:
  • 1
     Collective Intelligence
  • 2
     Generativity
  • 3
     Modularity
  • 4
     Bottleneck Control

55. A student says, "Collective intelligence just means the smartest person in the room leads everyone else." Which response best matches the study guide?
  • 1
     Correct, because collective intelligence mainly depends on the highest-IQ member.
  • 2
     Correct, because hierarchy is always the most intelligent group form.
  • 3
     Incorrect, because group performance can depend on interaction quality and shared contribution, not just one expert's ability.
  • 4
     Incorrect, but only in market-based decision systems.

56. In small-group tasks with measurable outcomes, what concept does the study guide use to describe group performance?
  • 1
     Network Effect
  • 2
     Ground Truth
  • 3
     C-factor
  • 4
     Platform Fit

57. For small-group tasks involving interaction, which factor is most strongly associated with higher C-factor according to the study guide?
  • 1
     The authority of the formal leader
  • 2
     The average IQ of the group
  • 3
     Members' social perceptiveness and more equal participation
  • 4
     The size of financial rewards

58. Which statement best reflects the study guide's idea that "diversity often trumps ability"?
  • 1
     Ability no longer matters at all if the crowd is large enough.
  • 2
     Diversity matters only in democratic voting systems.
  • 3
     Because knowledge is too widely distributed for one expert to master everything, diverse perspectives can sometimes outperform narrower expertise.
  • 4
     Diversity is mainly useful because it lowers wages and coordination costs.

59. Which Supermind type is correctly matched with its decision mechanism?
  • 1
     Hierarchy - mutual agreement among trading partners
  • 2
     Market - voting by equal members
  • 3
     Democracy - informal consensus and shared norms
  • 4
     Community - decisions shaped by informal consensus or shared norms

60. A stock market is used in the study guide as an example of which Supermind type?
  • 1
     Hierarchy
  • 2
     Market
  • 3
     Democracy
  • 4
     Community

61. A student council election is the clearest example of which Supermind type?
  • 1
     Hierarchy
  • 2
     Market
  • 3
     Democracy
  • 4
     Community

62. Which set best captures the three components of Collective Intelligence orchestration in the study guide?
  • 1
     Crowd size, platform revenue model, algorithmic pricing
  • 2
     Leader authority, incentive alignment, modular architecture
  • 3
     Nature of the crowd, motivation to contribute, and aggregation mechanism
  • 4
     Diversity score, voting speed, and data storage quality

파트8


63. A company is considering crowdsourcing a problem, but the crowd has no special insight beyond what the firm's own employees already know. Which Sweet Spot factor is most clearly weak?
  • 1
     IP Protection
  • 2
     Willing Crowd
  • 3
     Knowledge for Generating Solutions
  • 4
     Decomposability

64. A company believes the crowd may not be better at inventing solutions, but it could still judge which proposed solutions seem viable before production. Which Sweet Spot factor is strongest here?
  • 1
     Knowledge for Generating Solutions
  • 2
     Knowledge for Evaluating Solutions
  • 3
     IP Protection
  • 4
     Firm DNA

65. Which statement best matches the study guide's view of the five Sweet Spot factors?
  • 1
     All five factors must always be fully present for any kind of crowdsourcing to work.
  • 2
     Only willing crowd matters; the other factors are secondary.
  • 3
     Generation knowledge and evaluation knowledge do not always both need to be present, but the other three factors are still important.
  • 4
     If firm DNA supports crowdsourcing, the five factors are no longer necessary.

66. A firm wants to crowdsource a problem, but the problem cannot be separated from its internal proprietary workflow. Which Sweet Spot factor is the main obstacle?
  • 1
     Willing Crowd
  • 2
     Knowledge for Evaluating Solutions
  • 3
     Decomposability
  • 4
     Market Share

67. Which scenario best illustrates a failure of IP Protection?
  • 1
     A crowd is highly motivated by glory and love rather than money.
  • 2
     A crowd can clearly judge whether a design will appeal to consumers.
  • 3
     A problem can be broken apart and posted externally with little dependence on internal operations.
  • 4
     A firm can only explain the problem by revealing sensitive strategic information it does not want to leak.

68. Which statement about a "willing crowd" is most consistent with the study guide?
  • 1
     Contributors participate only for money; other motives are not durable enough.
  • 2
     A willing crowd matters only after production, not during idea generation or evaluation.
  • 3
     A crowd may be motivated by money, glory, or love, depending on the context and platform.
  • 4
     Willingness matters only when the problem involves no expertise at all.

69. A student says, "If a crowd can generate good ideas, it can always evaluate them well too." Which response best matches the study guide?
  • 1
     Correct, because generation and evaluation are effectively the same capability.
  • 2
     Correct, because crowdsourcing works only when the same crowd does both.
  • 3
     Incorrect, because generating solutions and evaluating solutions are different forms of knowledge and may not be equally strong in the same crowd.
  • 4
     Incorrect, but only for pro-social innovation problems.

70. Which example best shows that a problem may be suitable for crowds evaluating solutions even if crowds are not the main source of solutions?
  • 1
     A platform asks outsiders to vote on which design is most desirable before production.
  • 2
     A company hides all information and asks the crowd to guess the strategy.
  • 3
     A platform uses only employees to review internal proposals.
  • 4
     A crowd is invited only after the product is already launched.

71. Which statement best captures why firm DNA matters in the crowdsourcing decision?
  • 1
     Firm DNA replaces the need for decomposability and IP protection.
  • 2
     If the Sweet Spot factors fit, organizational beliefs no longer matter.
  • 3
     Even when a problem is structurally suitable for crowdsourcing, a firm may still resist it if its culture strongly favors internal R&D and internal control.
  • 4
     Firm DNA matters only for platform pricing, not for innovation choices.

72. Which statement best reflects the role of firm DNA in relation to the Sweet Spot framework?
  • 1
     Firm DNA is one of the five Sweet Spot factors and is always assessed first.
  • 2
     Firm DNA matters only when the crowd is unwilling to contribute.
  • 3
     Firm DNA and the Sweet Spot framework describe exactly the same idea from different angles.
  • 4
     Firm DNA is separate from the five-factor framework, but it can still influence whether a firm chooses to crowdsource.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
채점하기
hide
reset
타이머

모든 문제들의 저작권은 원저작권자에게 있습니다. 본 사이트는 웹상에 공개되어 있는 문제만 모아서 보여드립니다.
저작권 안내   데이터 보호 안내   제휴 문의

copyright 2026 뉴비티::새로운 CBT 시스템 - newbt.kr (Listed on LeanVibe)