Brainscape's Knowledge GenomeTM


Top Flashcards (Certified)
All Flashcards

MS IT-INTRO COMP > Gen AI > Flashcards

Gen AI Flashcards

1
Q

What is generative AI?

A

AI that can create original content such as text, images, video, audio or software code in response to a user’s prompt or request.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What are the three main phases of how generative AI operates?

A
  • Training
  • Tuning
  • Generation, evaluation and retuning
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is a foundation model in generative AI?

A

A deep learning model that serves as the basis for multiple different types of generative AI applications.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What are large language models (LLMs)?

A

Foundation models created for text generation applications.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What is the purpose of training in generative AI?

A

To create a foundation model by training a deep learning algorithm on huge volumes of raw, unstructured, unlabeled data.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

How does fine tuning differ from regular training?

A

Fine tuning involves feeding the model labeled data specific to the content generation application.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is reinforcement learning with human feedback (RLHF)?

A

A method where human users respond to generated content with evaluations to update the model for greater accuracy or relevance.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What does retrieval augmented generation (RAG) do?

A

Extends the foundation model to use relevant sources outside of the training data.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What are variational autoencoders (VAEs)?

A

Deep learning models that can encode data and decode multiple new variations of the content.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What is the role of generative adversarial networks (GANs)?

A

Comprise a generator that creates new content and a discriminator that evaluates the quality of the generated data.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is the main function of diffusion models?

A

To add noise to training data and then train the algorithm to iteratively diffuse the noise to reveal a desired output.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What is a transformer in the context of generative AI?

A

A deep learning model architecture that processes entire sequences of data and excels at natural language processing.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What types of content can generative AI create?

A
  • Text
  • Images and video
  • Sound, speech and music
  • Software code
  • Design and art
  • Simulations and synthetic data
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Fill in the blank: Generative AI can perform _______ tasks such as drafting summaries of documents.

A

repetitive

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

True or False: Generative AI can operate continuously without fatigue.

A

True

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What are some use cases for generative AI in enterprises?

A
  • Customer experience
  • Software development and application modernization
  • Digital labor
  • Science, engineering and research
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

What is a significant challenge of generative AI related to outputs?

A

Hallucinations and other inaccurate outputs.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

What does it mean for generative AI outputs to be inconsistent?

A

The same inputs can result in slightly or significantly different outputs.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

What is a major concern regarding bias in generative AI?

A

Generative models may learn societal biases present in the training data.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

What are ‘black box’ models?

A

Models where it is challenging to understand their decision-making processes.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

What risks does generative AI pose to security and privacy?

A

Can generate convincing phishing emails, fake identities or other malicious content.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

What are deepfakes?

A

AI-generated or AI-manipulated images.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

What are the potential malicious uses of generative AI models?

A

They can be exploited to generate convincing phishing emails, fake identities, or other malicious content.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

What is a deepfake?

A

AI-generated or AI-manipulated images, video, or audio intended to mislead people.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
What are the primary concerns associated with deepfakes?
They can damage reputations, spread misinformation, and are used in cyberattacks.
26
What is the significance of the term 'generative AI'?
It refers to AI that can create original content in response to user prompts.
27
List some key historical milestones in generative AI development.
* 1964: ELIZA, the first chatbot * 1999: Nvidia ships GeoForce * 2004: Google autocomplete * 2013: First variational autoencoders (VAEs) * 2014: First GANs and diffusion models * 2017: Transformer models introduced * 2022: Introduction of ChatGPT
28
What are the three phases of how generative AI operates?
* Training * Tuning * Generation, evaluation, and retuning
29
What is a foundation model in generative AI?
A deep learning model that serves as the basis for multiple generative AI applications.
30
What is the role of fine-tuning in generative AI?
It involves feeding the model labeled data specific to the content generation application.
31
What does RLHF stand for and what does it entail?
Reinforcement learning with human feedback; it involves human users evaluating generated content.
32
What is retrieval augmented generation (RAG)?
A framework for using relevant sources outside of training data to enhance a generative AI application's performance.
33
What are variational autoencoders (VAEs)?
Deep learning models that can encode data and generate multiple variations of content.
34
Describe the structure of a Generative Adversarial Network (GAN).
It consists of two neural networks: a generator and a discriminator.
35
What is the function of diffusion models?
They add noise to training data and then train the algorithm to iteratively remove the noise to reveal desired outputs.
36
What is the significance of transformers in generative AI?
They enable efficient training and high-quality content generation through attention mechanisms.
37
What types of content can generative AI create?
* Text * Images and video * Sound, speech, and music * Software code * Design and art * Simulations and synthetic data
38
How can generative AI enhance creativity?
By generating multiple novel versions of content to inspire creators.
39
What advantages does generative AI provide for decision-making?
It analyzes large datasets, identifies patterns, and generates hypotheses and recommendations.
40
What is the benefit of dynamic personalization in generative AI?
It generates personalized content in real-time based on user preferences.
41
What are some use cases for generative AI in customer experience?
* Drafting copy for marketing * Producing personalized marketing visuals * Powering next-generation chatbots
42
How does generative AI benefit software development?
By automating the process of writing new code and modernizing legacy applications.
43
What role does generative AI play in digital labor?
It can quickly generate or revise contracts, invoices, and other paperwork.
44
What potential does generative AI have in science and engineering?
It can help propose novel solutions to complex problems.
45
What is generative AI?
Generative AI is AI that can create original content such as text, images, video, audio or software code in response to a user’s prompt or request.
46
How does generative AI accelerate application modernization?
By automating much of the repetitive coding required to modernize legacy applications for hybrid cloud environments.
47
What role does generative AI play in digital labor?
It can quickly draw up or revise contracts, invoices, and other paperwork, allowing employees to focus on higher-level tasks.
48
In which fields can generative AI models assist?
* Science * Engineering * Healthcare * Research
49
What is an AI hallucination?
An AI hallucination is a generative AI output that is nonsensical or altogether inaccurate but seems plausible.
50
What preventative measures can developers implement to reduce AI hallucinations?
* Guardrails to restrict the model to relevant or trusted data sources * Continual evaluation and tuning
51
Why can generative AI outputs be inconsistent?
Due to the variational or probabilistic nature of generative AI models.
52
What is prompt engineering?
Iteratively refining or compounding prompts to achieve consistent results from generative AI applications.
53
What societal issue can generative models learn from training data?
Bias
54
How can developers prevent biased outputs in generative AI?
* Ensure diverse training data * Establish guidelines for preventing bias * Continually evaluate model outputs for bias
55
What does it mean that many generative AI models are 'black box' models?
It can be challenging or impossible to understand their decision-making processes.
56
What are deepfakes?
AI-generated or AI-manipulated images, video, or audio created to mislead people.
57
What malicious uses can deepfakes have?
* Damage reputations * Spread misinformation * Cyberattacks (e.g., voice phishing scams)
58
What historical milestone in generative AI occurred in 1964?
MIT computer scientist Joseph Weizenbaum develops ELIZA, the first chatbot.
59
What significant development in AI occurred in 2017?
The publication of 'Attention is All You Need,' documenting the principles of transformer models.
60
What was introduced by OpenAI in 2022?
ChatGPT, generating complex, coherent, and contextual content in response to prompts.
61
What is the significance of the Generative Pretrained Transformer (GPT)?
It represents a major advancement in large language models, with versions GPT-2 and GPT-3 rolled out in 2019-2020.
62
What do generative AI models rely on?
Sophisticated machine learning models called deep learning models.
63
What is a challenge in assessing the quality of generated content?
Traditional evaluation metrics may not capture creativity, coherence, or relevance.
64
True or False: Generative AI offers enormous productivity benefits for organizations.
True
65
Fill in the blank: Generative AI has made remarkable strides but presents significant ______.
challenges and risks
66
What must developers monitor to protect intellectual property (IP) when using generative AI?
Outputs for new content that exposes their own IP or violates others' IP protections.

MS IT-INTRO COMP (22 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2025 Bold Learning Solutions. Terms and Conditions