Quiz: Introduction to Generative AI
Test your knowledge about Generative AI and its fundamental concepts.
30 questions - Click on each question to reveal the correct answer.
Question 1
What is the main characteristic that distinguishes Generative AI from Traditional AI?
A) Generative AI is faster than traditional AI
B) Generative AI creates new content while traditional AI classifies or predicts
C) Generative AI uses less computing power
D) Generative AI requires no training data
Show Answer
Correct Answer: B
Generative AI creates new content (text, images, audio, etc.) based on learned patterns, while traditional AI analyzes, classifies, or makes predictions about existing data.
Question 2
When was the term "Artificial Intelligence" first coined?
A) 1943
B) 1950
C) 1956
D) 1969
Show Answer
Correct Answer: C
The term "Artificial Intelligence" was coined at the Dartmouth Conference in 1956, which is considered the birth of AI as a field.
Question 3
What is a "prompt" in the context of Generative AI?
A) A command to shut down the AI
B) A text instruction that guides the AI's output
C) A type of training data
D) An error message from the AI
Show Answer
Correct Answer: B
A prompt is a text instruction provided by the user that guides the AI in generating specific content. The quality of the output often depends on the quality of the prompt.
Question 4
Which of the following is NOT a type of content that Generative AI can create?
A) Text
B) Physical objects
C) Images
D) Audio
Show Answer
Correct Answer: B
Generative AI can create digital content like text, images, audio, video, and code, but it cannot directly create physical objects. It can only design or describe them.
Question 5
Who proposed the Turing Test?
A) Geoffrey Hinton
B) Alan Turing
C) Sam Altman
D) Ian Goodfellow
Show Answer
Correct Answer: B
Alan Turing proposed the Turing Test in 1950 as a way to evaluate a machine's ability to exhibit intelligent behavior indistinguishable from a human.
Question 6
What was ChatGPT's achievement regarding user adoption?
A) First AI to pass the Turing Test
B) Fastest-growing consumer application in history
C) First AI to generate images
D) First open-source AI model
Show Answer
Correct Answer: B
ChatGPT reached 100 million users in just 2 months, making it the fastest-growing consumer application in history.
Question 7
What does GAN stand for?
A) General Artificial Network
B) Generative Adversarial Network
C) Global AI Network
D) Generated Automatic Node
Show Answer
Correct Answer: B
GAN stands for Generative Adversarial Network, invented by Ian Goodfellow in 2014. It consists of two neural networks (generator and discriminator) that compete to produce realistic content.
Question 8
Which year marked the introduction of the Transformer architecture?
A) 2014
B) 2015
C) 2017
D) 2020
Show Answer
Correct Answer: C
The Transformer architecture was introduced in 2017 in the paper "Attention Is All You Need" by Google researchers. It became the foundation for all modern large language models.
Question 9
What is "hallucination" in the context of Generative AI?
A) A visualization technique
B) Generating plausible but incorrect information
C) A training method
D) A type of prompt engineering
Show Answer
Correct Answer: B
Hallucination refers to when AI generates content that sounds plausible but is factually incorrect or made up. This is a known limitation of generative AI systems.
Question 10
Which of these is an example of traditional (discriminative) AI?
A) ChatGPT writing an email
B) DALL-E creating an image
C) A spam filter classifying emails
D) Suno generating music
Show Answer
Correct Answer: C
A spam filter classifying emails as spam or not spam is an example of traditional discriminative AI. It analyzes and classifies existing content rather than creating new content.
Question 11
What does LLM stand for?
A) Large Learning Model
B) Large Language Model
C) Limited Language Machine
D) Linear Learning Method
Show Answer
Correct Answer: B
LLM stands for Large Language Model. These are AI models trained on massive amounts of text data that can understand and generate human-like text.
Question 12
What was ELIZA (1966)?
A) One of the first chatbots
B) An image generator
C) A chess-playing AI
D) A translation tool
Show Answer
Correct Answer: A
ELIZA was one of the first chatbots, created in 1966 at MIT. It simulated a psychotherapist and could engage in conversation using pattern matching.
Question 13
What is the key innovation of the self-attention mechanism in Transformers?
A) It uses less memory
B) It considers relationships between all words simultaneously
C) It only processes one word at a time
D) It requires no training
Show Answer
Correct Answer: B
The self-attention mechanism allows the model to consider relationships between all words in a sentence simultaneously, regardless of their distance from each other, enabling better understanding of context.
Question 14
When was ChatGPT released?
A) January 2022
B) November 2022
C) March 2023
D) December 2021
Show Answer
Correct Answer: B
ChatGPT was released on November 30, 2022, and quickly became a global phenomenon.
Question 15
Which company created GPT and ChatGPT?
A) Google
B) Meta
C) OpenAI
D) Microsoft
Show Answer
Correct Answer: C
OpenAI created GPT (Generative Pre-trained Transformer) and ChatGPT. Microsoft is a major investor in OpenAI but didn't create these models.
Question 16
What are "AI Winters"?
A) Cold server rooms for AI
B) Periods of reduced funding and interest in AI
C) Training data collection periods
D) Winter-themed AI applications
Show Answer
Correct Answer: B
AI Winters were periods of reduced funding, interest, and progress in AI research, typically following periods of overhyped expectations that weren't met. Major AI winters occurred in the 1970s and late 1980s.
Question 17
Which of the following best describes the role of the Discriminator in a GAN?
A) Creates fake content
B) Tries to distinguish between real and fake content
C) Stores the training data
D) Compresses the images
Show Answer
Correct Answer: B
In a GAN, the Discriminator tries to distinguish between real content (from the training data) and fake content (created by the Generator). This adversarial process helps both networks improve.
Question 18
What is the relationship between Deep Learning and Generative AI?
A) They are the same thing
B) Deep Learning is a subset of Generative AI
C) Generative AI often uses Deep Learning techniques
D) They are completely unrelated
Show Answer
Correct Answer: C
Generative AI often uses Deep Learning techniques, particularly neural networks and the Transformer architecture. Deep Learning provides the foundation for most modern generative AI systems.
Question 19
Which image generation tool is known for being open-source?
A) DALL-E
B) Midjourney
C) Stable Diffusion
D) Adobe Firefly
Show Answer
Correct Answer: C
Stable Diffusion is known for being open-source, allowing anyone to download, modify, and run the model locally. DALL-E and Midjourney are proprietary services.
Question 20
What does it mean that Generative AI is "probabilistic"?
A) It always gives the same output
B) The same prompt can produce different outputs
C) It only works with probability calculations
D) It cannot make decisions
Show Answer
Correct Answer: B
Being probabilistic means the AI makes choices based on probability distributions, so the same prompt can produce different outputs each time. This introduces variety but also unpredictability.
Question 21
Who is known as the "Godfather of Deep Learning"?
A) Alan Turing
B) Geoffrey Hinton
C) Sam Altman
D) Elon Musk
Show Answer
Correct Answer: B
Geoffrey Hinton is often called the "Godfather of Deep Learning" for his pioneering work on neural networks. He won the Nobel Prize in Physics in 2024 for his contributions to machine learning.
Question 22
What capability did GPT-3 demonstrate that surprised researchers?
A) Ability to fly drones
B) Emergent abilities like writing code without explicit training
C) Perfect memory of all training data
D) Self-awareness
Show Answer
Correct Answer: B
GPT-3 demonstrated emergent capabilities—abilities that weren't explicitly trained for, such as writing code, creating poetry, and translation. These abilities "emerged" from the scale of training.
Question 23
Which company created Claude, the AI assistant focused on safety?
A) OpenAI
B) Google
C) Anthropic
D) Microsoft
Show Answer
Correct Answer: C
Anthropic created Claude, with a focus on AI safety and alignment. Anthropic was founded by former OpenAI researchers including Dario Amodei.
Question 24
When should you prefer Traditional AI over Generative AI?
A) When you need to create new content
B) When you need a specific classification or prediction
C) When you want creative brainstorming
D) When writing blog posts
Show Answer
Correct Answer: B
Traditional AI is preferred when you need specific classifications (spam/not spam) or predictions (sales forecast) rather than content creation. It's often more accurate and efficient for these structured tasks.
Question 25
What is the main purpose of training data for Generative AI?
A) To store the AI's responses
B) To teach the AI patterns and structures in content
C) To test the AI's performance
D) To limit what the AI can say
Show Answer
Correct Answer: B
Training data teaches the AI patterns, structures, and relationships in content. The AI learns from this data to generate new, similar content based on these patterns.
Question 26
What event in 1997 demonstrated AI could master complex games?
A) AlphaGo defeating a Go champion
B) Deep Blue defeating Garry Kasparov
C) Watson winning Jeopardy
D) GPT playing chess
Show Answer
Correct Answer: B
In 1997, IBM's Deep Blue defeated world chess champion Garry Kasparov, demonstrating that AI could master complex strategic games. AlphaGo's Go victory came much later in 2016.
Question 27
Which of the following is a characteristic of Generative AI outputs?
A) Always 100% accurate
B) Novel content based on learned patterns
C) Exact copies of training data
D) Completely random content
Show Answer
Correct Answer: B
Generative AI creates novel content that didn't exist before, based on patterns learned from training data. It's not copying, not random, and not always accurate.
Question 28
What does "multimodal" mean in the context of AI?
A) AI that only processes text
B) AI that can work with multiple types of content (text, images, audio)
C) AI that runs on multiple computers
D) AI that speaks multiple languages
Show Answer
Correct Answer: B
Multimodal AI can understand and generate multiple types of content—such as text, images, and audio—often in combination. GPT-4 with vision capabilities is an example of multimodal AI.
Question 29
Why is human oversight still important when using Generative AI?
A) AI is always wrong
B) Legal requirements only
C) AI can hallucinate and make errors
D) AI needs someone to press buttons
Show Answer
Correct Answer: C
Human oversight is essential because AI can hallucinate (generate false information), make errors, and lacks true understanding. Humans need to verify AI outputs, especially for critical applications.
Question 30
What is the significance of the "Attention Is All You Need" paper?
A) It introduced the first neural network
B) It introduced the Transformer architecture that powers modern LLMs
C) It created ChatGPT
D) It proved AI is conscious
Show Answer
Correct Answer: B
The 2017 paper "Attention Is All You Need" introduced the Transformer architecture, which became the foundation for all modern large language models including GPT, BERT, and Claude.
Results
| Score | Level |
|---|---|
| 27-30 | Excellent - Strong understanding of Generative AI fundamentals |
| 21-26 | Good - Some review may be helpful |
| 15-20 | Average - Review the chapters again |
| Below 15 | Needs comprehensive review |