IT COMMUNITY

Introduction to Generative AI

What is Generative AI?

Final Exam:
1. In considering the future implications of generative AI on the job market,
what skill set is essential for individuals seeking to thrive in this evolving
landscape?
Ans. development of unique personal emotional, interpersonal, and creative
abilities
2. What are the two primary recommendations for an executive or leader to
implement with AI?
Ans. self-questioning and a moral compass
3. What is the most notable functionality of Natural Language models like
ChatGPT?
Ans. Its large scale capability to generate human-like text
4. You are an experienced AI user and want to be able to choose different
models. What will you use?
Ans. A notebook
5. How should we think of the relationship between humans and generative
AI?
Ans. Human input and creativity will work in conjunction with Al to produce
meaningful progress.
6. What is true about using text-to-image generation services?
Ans. They are dependent on dataset quality.
7. Which AI focuses on classifying or identifying content that is based on
preexisting data?
Ans. discriminative.
 8. Why might marketers use text-to-image models in the creative process?
Ans. They can be more efficient to use and provide a unique look and feel.
9. Do you need to have a technical AI background in order to start a
generative AI venture?
Ans. No. As a business leader with no technical background, you can either
make partnerships with generative Al research institutions, or you can use
open-source models in your next business endeavor. And as a hobbyist, a
maker, or a creative, you can make use of a generative Al services that are free
or are accessible with a modest fee.
10. How can Variational Autoencoders (VAEs) be used in anomaly detection?
Ans. VAEs can be trained on a dataset of normal data, and later on be used to
identify instances that deviate from the normal data.

Generative AI vs. Traditional AI

Final Exam:
1. Imagine that you work for a health insurance company that wants to use a
foundation model to approve medical procedures for your customers. One of
the product managers mentions that if a customer questions a decision, how
will customer service representatives answer these concerns? The data
scientist mentions that the system will use 10s of thousands of variables that
might be difficult to explain. How would you approach this ethical challenge?

Ans. Try to determine if there's any way to make the decisions traceable.

2. What’s a good definition for generative AI?

Ans. Generative AI looks at existing data patterns to create something completely

new.

3. You work for an organization that’s creating a virtual guidance counselor for students
who are graduating from high school. Your organization created a foundation model
that went through huge amounts of data on travel, occupations, and colleges. In
testing, you've found that your virtual guidance counselor will often give students
misinformation. It will exaggerate salaries for careers and give inaccurate information
about college admission requirements. How would you approach this challenge?

Ans. Generative AI systems currently tend to produce hallucinations. You can try
to limit this by making sure the model uses quality data.
4. OpenAI uses a model called GPT, which stands for generative pre-trained
transformer. This is the model they used for their ChatGPT large language model
(LLM) chatbot. What does it mean that this model is “pre-trained?”

Ans. It means that the LLM looks for patterns, so it can guess the next word in a
sentence.

5. In 2011, the Google brain project produced ghostly images from patterns that I
identified in YouTube videos. The deep learning network used unsupervised learning,
so it didn't label these patterns. But when Google engineers went through and looked
at what the network was seeing they were able to identify cats and people's faces.
What impact did this discovery have on future systems?

Ans. It showed the potential for using this technology to generate new images
from massive datasets.

6. Imagine that you work as a book lithographer. The illustrations that you create
are heavily sought after by textbooks, and other artistic publishers that
produce high quality books. Now a company that creates illustrations for
books has trained their generative AI system to copy your technique. What
used to take you months can now be produced by this system in seconds at a
very low cost. You've had a noticeable drop in commissioned works, and you
had to fire many young artists. What type of impact will this have on future
artists?

Ans. If work decreases, you cannot train upcoming artists to generate new art for
training future GenAI systems.

7. Imagine that you work for a digital music service. Your company finds that
people are more likely to listen to music when you make recommendations.
So your data scientists want to create an artificial neural network that will
cluster together songs that have similar characteristics. Your company has
over 100 million songs that could be clustered together regardless of the
genre. What type of machine learning would be the best fit?

Ans. unsupervised learning

8. Imagine that you're a movie critic for a newspaper in the United States. You've
written a column for decades. You've also written several books. In fact, many
subscribers say that your column is one of the reasons that they subscribe.
Recently you've been thinking about leaving the paper. The newspaper found
out about this, and so they decided to use a generative AI system to model all
your previous columns and books. They can create your movie column
without you. Under current law, has the newspaper done anything illegal?
 Ans. No, under current law the newspaper could train a model on your columns
and books under the fair use doctrine.

9. You work for an organization that wants to generate images of customers

wearing new outfits. The data scientists in your organization are trying to
decide whether your system should use a diffusion model or a generative
adversarial network (GAN). Some data scientists say not to use GANs because
they focus more on creating different outfit compositions with the customer's
image rather than generating a photorealistic image. Do you think they're
right?

Ans. Yes, a diffusion model would probably work better because the clarity isn't as
important as the composition.

10. What is one of the main challenges that self-supervised learning is trying to
solve?

Ans. There's a lot more unlabeled data available, so these systems use
unsupervised learning to label data for foundation models.

11. Imagine that you're looking at a picture of a male lion sitting peacefully in a
grassy savannah. If you used a variational autoencoder (VAE) to analyze this
image, what might be some of the items that get captured by anomaly
detection?

Ans. Any of the items that are “non-lion,” this could include the grass, the trees,
and the sky.

Generative AI in Learning and Development

Final Exam:
1. How can GenAI best help increase the shelf life of an organization's learning
material?
Ans. by helping L&D leaders continuously analyze and update learning material
2. How can GenAI tools best help you as an L&D leader?
Ans. by increasing work productivity and efficiency.
3. According to a recent PwC Global survey, what percentage of CEOs were
reported to be concerned about a lack of skills across their organizations?
Ans. 70%
 4. How can GenAI best help you as an L&D leader address individual learning
needs?
Ans. by creating personalized learning paths

5. Tamara has been developing use cases for GenAI at her organization. How
should she proceed after defining the objectives and outcomes for her use
cases?
Ans. Assess potential risks and create mitigation strategies.

6. How can GenAI best assist you in demonstrating the results of your
organization's leadership development programs?
Ans. It can analyze program and leadership performance data.

7. Why can GenAI have some amount of inherent bias?

Ans. It is trained on data from the internet.

Integrating generative ai into business strategy

FINAL EXAM:
1. Paalavi has an inventory of AI use cases and understands the requirements
and risks involved. What is the next step she must take?
Ans. Create an AI roadmap.

2. What is the primary output of conducting an AI feasibility assessment?

Ans. a technical requirements document

3. An organization is starting the AI journey with its unique environment. In

which USAGE phase will the organization focus on the prototyping and testing
of its business scenarios?
Ans. Guide

4. You are thinking about AI literacy at a strategic level. Which factor is needed
first to empower employees to improve business operations?

Ans. AI is a business advantage.

 5. Your boss appoints you as a dedicated AI leader. What is the next step in
structuring AI leadership and ownership for the business?

Ans. Implement cross-functional AI committees to act as communication

conduits.

6. Janine is using the USAGE framework. She is now at the Guide stage where
she will take the AI strategy to execution. Which approach is least practical at
this point?

Ans. Start large in scope and work toward smaller solutions.

7. Terrance is conducting an AI skills gap analysis. What should he do first to

address any skills gaps?
Ans. Look internally for resources.

8. In what stage of the USAGE framework is it necessary to develop an agile

roadmap to drive short- and long-term value?
Ans. Align
9. What is the recommended first pillar that you should address to form the
foundation of your AI action plan?
Ans. Prioritize data integrity and security.

How to Research and Write Using Generative AI Tools

Final Exam:
1. You are using ChatGPT to write a sales pitch for a new washing machine.
Using the CREATE approach, what should you do in the "request" aspect of
your prompt in order to get the best output?
Ans. Be as specific and detailed as possible.

2. In analyzing the tone of the piece of writing, what can ChatGPT evaluate?
Ans. seriousness vs. humor

3. Why is generative AI a significant new development in computing?

Ans. It can augment human thinking to help individuals achieve more than they
could without AI.

4. You have been tasked with acquiring high-quality information for a project.
What is the most effective way to use ChatGPT in this situation?
Ans. Ask ChatGPT to suggest queries to use with internet search engines.
5. Jamal wants to create a strategic business model for the consulting business
that he is starting. How can he best use generative AI tools to help him?
Ans. by asking AI to provide final step-by-step instructions for a SWOT analysis

6. Malik, who works in user design and marketing, used generative AI to create
seven personas for a new software program. What is TRUE about these
personas?
Ans. They are likely to be helpful in product marketing.

7. You are researching 18th century poetry using ChatGPT. After getting an initial
response, you prompt, "Provide analogies for your comparison of Blake and
Burns." How is ChatGPT likely to respond?

Ans. by generating analogies as requested

8. Jen has asked ChatGPT to create a table showing when various computer
chips came into use, how long they were used, and for what purpose. What
will ChatGPT likely do in response to her prompt?

Ans. Produce a detailed table showing the information Jen has requested

9. What is the best way to use generative AI to improve a piece of writing?

Ans. Request that your copy be rewritten to stylistically match that of a nationally
known syndicated columnist.

10. What is true about book summaries generated by ChatGPT or similar AI tools?

Ans. Although generally accurate, the summaries sometimes include fabricated

statements.

11. While using ChatGPT, you write a prompt that says, in part, "Start by asking
me questions to gather information. Do not offer a response until you have
my answers." How will ChatGPT most likely reply?

Ans. by asking you questions as prompted

12. In optimal interactions between humans and AI, what role should AI play?

Ans. collaborator
13. What is true about prompts for ChatGPT or other generative AI tools?
 Ans. Prompts should contain a fair amount of detail about the audience, the
desired output, and the hypothetical author of the content.

Generative AI for Business Leaders

Final Exam:
1. Clete is considering AI for his company's data analytics, but he is concerned
about data security. How can AI most easily help him?
Ans. AI can generate synthetic data that mimics his company's without exposing
its real data.

2. Ivan works for a small company and is giving a presentation to management

on the coming advancements in generative AI. Where should he begin in
order to gain the managers' interest?
Ans. Explain that continually decreasing costs will make AI more available for
their company.

3. An engineering company is anticipating an explosive increase in sales from an

AI-generated model that produces a unique manufacturing software. What
must this company consider the most?
Ans. scalability
4. If a company decides to start using AI, what must be the starting point?
Ans. what its business objectives are

5. How can you best think of AI?

Ans. as an intelligent assistant working beside you on everything you are trying to
accomplish

6. Sean always brings unique and nuanced input into his prompts. What
limitation of AI is he addressing?

Ans. token constraints

7. WorldCars LTD has adopted generative AI for its sports car manufacturing
process. What is the most significant financial advantage that this business
can derive?

Ans. The AI system can model assembly-line setups before they are implemented
in the real world.
8. SalesCo wants to use AI for its customer support. What is the best approach
for the company to use?

Ans. Adopt the latest AI models that are trained on best practices for every
product knowledge base available.

9. May is creating product bulletins for customers and salespeople. The

customer-facing bulletin should highlight benefits, and the sales-facing one
should give specifications. How should May do this?

Ans. Provide the AI model the proper context, and then design two separate
nuanced prompts.

10. What is the most important consideration for a company looking for a partner
or vendor to help it implement an AI system?

Ans. knowing what the company needs

Introduction to Generative AI with GPT

Final Exam:
1. When was GPT first introduced to the world?

Ans. in 2018

2. What GPT capability is available with images?

Ans. The ability for GPT to describe the contents of an image

3. What is one of the top five AI skills for IT professionals today that CompTIA
identified?

Ans. support and troubleshooting of AI issues

4. What is the central feature of GPT?

Ans. creating output in the form of human-like text

5. Should companies consider greater collaboration between their business and

IT portions as AI becomes more important?
 Ans. Yes, both groups will need to be involved to achieve success in an
increasingly competitive market.

6. Which area is not normally considered for machine ethics?

Ans. potential profits

7. Why is it important for leaders and managers to understand what is possible

with AI?

Ans. to define a vision for AI in their organization

8. Who were two of the founders of OpenAI?

Ans. Elon Musk and Sam Altman

9. Why is today's current AI considered "narrow AI"?

Ans. It performs certain tasks well.

10. What does GPT stand for?

Ans. Generative Pre-trained Transformer

11. Which of these is a serious issue with GPT?

Ans. Output can be incorrect.

COMMUNITY LINK
https://chat.whatsapp.com/LeCPQMcHu787cATgR5
B0jw