Skip to content

Topic Drill 1.1.1: AI-Based and Conventional Systems ​

About this drill

LO AI-1.1.1 (K2): Differentiate between AI-based systems and conventional systems.
Lesson: Ch 1, section 1.1.1

This drill covers every point in section 1.1.1:

  • How behavior is defined: imperative, step-by-step instructions vs. patterns learned from data (the cat example)
  • Problem-solving: deterministic and transparent vs. probabilistic reasoning, statistical inference and pattern recognition
  • Explainability: the black-box problem, critical domains, and regulation's focus on transparency and explainability
  • Adaptability: static systems with manual updates vs. self-learning systems, and the need for continuous monitoring

Topic Drill: 1.1.1

Q1. A bank replaces its hand-coded credit rules (if-then-else checks on income bands) with an ML model trained on ten years of loan outcomes. Which change should the test team expect MOST?

AExpected results become easier to derive, because the model was trained on real historical decisions
BThe model will reproduce the old rules exactly, because it learned from decisions made with those rules
CExpected results can no longer be derived from explicit rules, because the behavior comes from patterns learned from data
DThe system becomes fully deterministic, because the training data is fixed

Q2. Which property of conventional systems makes it easier for humans to understand how inputs produce outputs?

ADevelopers explicitly define step-by-step instructions, which makes the behavior deterministic
BThey rely on statistical inference to choose the most likely output
CThey continuously improve as they encounter new data
DThey recognize patterns in large datasets

Q3. A hospital evaluates a diagnosis aid built on a deep learning model with billions of parameters. The regulator asks the hospital to show WHY the system recommended a particular treatment. Which characteristic of AI-based systems causes the difficulty?

AAdaptability
BLimited explainability ("black-box" nature)
CDeterministic behavior
DImperative programming

Q4. Which TWO domains does the syllabus name as critical domains where understanding why an AI model made a decision is crucial?

Select 2 options.

AHealthcare
BOnline gaming
CFinance
DMusic streaming
ESocial media advertising

Q5. A logistics company deploys a self-learning routing system that keeps improving from new traffic data. Which statement BEST reflects the syllabus view of this adaptability?

AIt removes the need for regression testing, because the system only improves
BIt means the system must be manually updated whenever traffic patterns change
CIt makes the system easier to explain, because it learns from real data
DIt makes the system powerful in dynamic environments, but requires continuous monitoring to stay aligned with its core requirements

Q6. Why are many AI-based systems able to handle complex forms of uncertainty and ambiguity more effectively than conventional systems?

ABecause they are written in imperative languages with exhaustive if-then-else statements
BBecause they are always trained on complete and unambiguous data
CBecause they rely on probabilistic reasoning, statistical inference and pattern recognition
DBecause their outputs are always identical for the same inputs

Q7. A rule-based pricing engine must be changed by developers whenever market conditions change. What does this illustrate?

AThe self-learning nature of AI-based systems
BConventional systems are static and typically require manual updates to incorporate new knowledge
CThe black-box nature of conventional systems
DThe probabilistic nature of conventional systems

Q8. Which statement about AI-based and conventional systems is LEAST accurate?

AConventional systems make it easier to understand how inputs produce outputs
BAchieving transparency and explainability has become a critical focus in AI regulation
CAn AI image recognizer for cats learns from a dataset of cat images rather than from coded rules
DBecause they are trained on fixed data, AI-based systems always produce predictable outputs

0 / 8 answered

Next drill: 1.1.2 Narrow AI, General AI, and Super AI →

Unofficial study guide for the ISTQB® Certified Tester AI Testing (CT-AI) exam. Not affiliated with International Software Testing Qualifications Board (ISTQB®).