Google AI Essentials

Certification: https://coursera.org/share/3caa5803f5adcfc2fdadfab2f4ffede6

https://www.coursera.org/learn/google-introduction-to-ai#modules course (upto 4h) | https://www.coursera.org/learn/google-introduction-to-ai#modules

The TCREI framework is a 5-step method for crafting effective AI prompts, standing for Task, Context, References, Evaluate, and Iterate,
- ex:
Task: Create a social media post about an upcoming music festival that speaks to the local music community while attracting out-of-state festival-goers.

Persona: You're a concert promoter specializing in raising ticket sales in the alternative rock music industry.

Format: Limit the post to 125-characters. Include 5 relevant hashtags.

Module 1: Introduction to AI

Overview of AI tools

Our aim is to provide you with fundamental AI skills—regardless of the platform. While you are welcome to use any AI tool you are comfortable with, our activities feature these tools:

Gemini is a versatile AI tool designed for a wide range of tasks, such as brainstorming ideas, summarizing complex documents, writing code, and even analyzing the content of images. You can also use these powerful features in the tools you use every day with Gemini in Google Workspace. Gemini's flexibility makes it ideal for practicing the AI concepts you'll learn in this specialization. Learn more about Gemini's features and regional availability.
NotebookLM is a research and writing assistant that is grounded exclusively in the source materials you provide, instead of generating information from other sources. Learn more about NotebookLM's features and regional availability.
AI Studio is a web-based tool designed for prototyping and experimenting with AI models. Learn more about AI Studio's features and its regional availability
Streaming Service Recommendations by AI: An AI tool refers to AI-powered software that can automate or assist users with a variety of tasks. Examples of AI tools are everywhere, from GPS systems that suggest quick routes to translation systems that interpret conversations in real time.
AI Tool
machine learning: Machine learning, or ML, is a subset of AI focused on developing computer programs that can analyze data to make decisions or predictions
- AI designers build ML programs using a training set, which is a collection of data used to teach AI.
- However, for ML programs to perform effectively, the quality and relevance of their training data matter. A fundamental issue to be aware of is the potential for bias within training data. This could unintentionally cause an AI tool to produce inaccurate or unintended outputs.
  - For example, the AI tool that was used to sort ripe apples might have learned from training data that only contain images of specific types of red apples. This would unintentionally make the AI less accurate at identifying ripe apples of varying sizes, shapes, or colors. The food producer might end up sorting apples incorrectly, causing them to lose money and waste perfectly good apples.
generative AI: generative AI is AI that can generate new content such as text, images, or other media.
- A unique quality of generative AI tools is that you can use them with natural language.
natural language: Natural language refers to the way people talk or write when communicating with each other
- How a generative AI tool works with natural language:
  1. Provide Input: Input refers to any information or data that's sent to a computer for processing. Many generative AI tools, accept text and speeches input, and some also accept images or video files.
  2. Data is Processed: Next, the data is processed by the AI tool.
  3. Output is Generated: in the form of text, images, audio, or video.
Benefits of generative AI
- Boost your productivity
- Help you avoid mistakes
- Improve your decision-making process
Conversational AI Tool: A conversational AI tool is a generative AI tool that processes text requests and generates text responses.
- You can use it to brainstorm ideas, answer questions, and boost your productivity.

A Guide to AI and ML

Rule-based AI: Rule-based AI operates on hard-coded rules set by developers and does not learn from new data, making it less flexible than machine learning approaches.
Reinforcement Learning: Reinforcement learning is a method where tools learn through trial-and-error, guided by feedback, to improve their performance on specific tasks.
Machine Learning (ML): Machine learning is a subset of AI that enables tools to learn from data without being explicitly programmed for every scenario, allowing for adaptive and intelligent behavior.
Generative AI: Generative AI refers to tools that can create new and original content, such as text, images, music, and code, using machine learning techniques.
Unsupervised Learning: Unsupervised learning involves training tools on unlabeled datasets to identify patterns and structures in data without predefined outcomes.
Supervised Learning: Supervised learning is a machine learning approach where tools are trained on labeled datasets to predict specific outcomes, such as recognizing features in images.

Generative AI and machine learning

Artificial intelligence is a broad field focused on creating tools that can complete tasks typically associated with human intelligence. Generative AI is a type of AI, and it is powered by machine learning (ML), a specific technique that enables a tool to learn from data without being explicitly programmed for every possible scenario. While both terms seem similar, ML is a subset of AI used by many of the tools available today.

Approaches to machine learning

There are three common ML approaches used to develop AI tools:

Supervised learning, Unsupervised learning, Reinforcement learning

Supervised learning is used to train tools from a massive dataset that has been labeled by humans. This technique is often used when there is a specific, known output in mind. For example, an image generator is trained on millions of labeled pictures, like ones explicitly labeled "cat." ML enables the tool to recognize the features, patterns, and characteristics of cats so that it can create custom, new images.
Unsupervised learning is used to train tools from a dataset that has not been labeled by humans. This technique is used to identify patterns and structures in data when there isn't a specific, known output in mind. For example, an image generator analyzes a large dataset of animal photos. ML enables the tool to identify patterns on its own, clustering images with similar features like whiskers and pointy ears. This allows it to learn what a "cat" looks like without any human-provided labels.
Reinforcement learning is used to train tools through a process of trial-and-error that is guided by feedback. This technique is used to continuously refine and improve a tool's performance on a specific task. For example, after an image generator creates a picture of a cat, it receives feedback from a human evaluator. If the feedback is positive, this signals that the output was successful. This feedback is collected and used by developers to help improve future versions of the AI tool.

Many of today's AI tools use a combination of all three ML approaches to create text, images, video, and more.

However, it's important to note that the "learning" described in these approaches only happens during the tool's development and training—before it's released to the public. The feedback and data collected from users helps developers improve future versions of the tool, but the AI is not actively learning in real-time as you use it. You'll learn more about the specifics of the AI training process later.

Rule-based AI: A different approach

While many modern AI tools are powered by machine learning, another common approach is rule-based AI. These tools operate using a set of hard-coded rules created by human developers and do not learn from new data. They follow their specific instructions precisely.

For example, a simple customer service chatbot might be programmed with the rule: "If a user's message contains the phrase 'tracking number,' respond with a link to our package tracking website." The chatbot will follow this rule every time, but it cannot adapt or understand any requests that fall outside of its pre-written rules.

You may encounter rule-based AI in workplace tools that are designed for predictable tasks. This approach is less flexible than machine learning, which allows AI tools to adapt and handle a wider range of complex, real-world data.

Resources for more information

If you'd like to learn more, check out PAIR Explorables, a collection of interactive articles that can help you explore different AI concepts and experience how they work.

Understand the capabilities and limitations of AI

Capabilities of AI

Today's AI tools can do a lot to enhance your work. They can generate content, like assisting a marketing team by making a promotional video for a new product. They can analyze information quickly, like highlighting the key points of a long email thread. They can answer questions in a detailed and nuanced way. And overall, they can simplify your day-to-day and allow you to focus on other aspects of your work.

Limitations of AI

While AI can complete a variety of tasks, there are some tasks that require a human touch, such as handling sensitive issues. These limitations can be critical in certain contexts. For example, AI can't learn independently. It needs people to continually update its training. Shortcomings in an AI tool's training data can also potentially reflect or amplify biases, leading to skewed or unfair outcomes. Another major limitation is that AI output can sometimes contain inaccuracies, otherwise known as hallucinations.

Hallucinations: Hallucinations are AI outputs that are not true.

These inaccuracies can range from minor errors, such as a sentence that doesn't make sense, to significant distortions. For instance, consider a sales manager who's using an AI tool to analyze quarterly sales data. The AI tool might identify declining sales of a particular product, and flag the item as something that should be removed from stores. However, what if there were a seasonal factor affecting sales that hadn't been accounted for in the AI tool's analysis? Hallucinations like this one can lead to misguided decisions if the user doesn't carefully review the AI tool's output.
Considering AI's limitations, human oversight over AI generated output is crucial to ensure that the information is accurate and ethical.

Activity: Use AI to create an email

AI tool access

To complete this activity, you can use any browser-based generative AI tool of your choice. Generative AI technology is dynamic and ever-changing, so always double check what features and capabilities are available in the tools you’re using. To use Gemini for this activity, go to gemini.google.com.

Remember that providing human oversight is crucial when using an AI tool, so always be sure to review AI-generated output and give additional instructions to the tool as needed.

Scenario: Draft an email

Think of a scenario where you need to send an email. It could be a progress update on a project or a simple note thanking your team for their work on a recent task.

Once you’ve thought of a scenario, you'll use generative AI to help you draft the email. After that, you’ll evaluate and revise it to make sure it meets your needs.

Step 1: Prompt the AI tool to draft an email

In your browser, open a new tab to search for your generative AI tool of choice, and begin prompting.

As a reminder, a prompt is text input that provides instructions to an AI tool on how to generate output. Prompts can take a variety of different forms, and you can phrase a prompt in any way that feels natural to you. When prompting, just make sure you provide clear and specific instructions and include any relevant details.

Here are some examples of prompts you can use to create an email:

I'm a project manager and need to write a thank you email to my colleagues. I want to show my appreciation for the team's hard work on a recent project. Help me draft a short email that will be sent to all team members.
Write an email to my coworker, who is currently working on an important quarterly project. Ask for a status update on the project, and offer my assistance with any tasks that they are behind on.

Now think of how you can prompt an AI tool to create an email for your scenario. When you're ready, enter your prompt into the tool.

Step 2: Review the output

Review the email that the AI tool outputs, and evaluate whether it meets your needs.

AI tools might not generate exactly what you need on the first try, even with a clear initial prompt. This isn't necessarily because the tool made a mistake—sometimes you might realize you want a different tone, more context, or a change of style. For example, the email might include a sentence that isn’t relevant, or you might get a technical response that needs more of a personal tone.

Now that you've reviewed the output, consider some ways the email could be improved. Keep those ideas top-of-mind as you move on to the next step.

Step 3: Revise the Email

Now that you’ve reviewed the email, think of how you can improve it, and continue prompting.

To refine AI-generated output, engage in a back-and-forth conversation with the AI tool. It might take multiple rounds of feedback to get the output you need. For example, if you feel the email is too long, ask the AI tool to make it more concise, or if an important detail is missing, ask the AI tool to include it.

Prompt as many times as needed until the email is ready.

Use AI as a collaborative tool

AI augmentation: AI augmentation refers to the process of using AI to improve a work product, whether by making it easier to do or higher in quality.
AI automation: AI Automation refers to the process of using AI to accomplish tasks without any action on the user's part.

Quiz

1:a, 2: a,c,d 3: a, d 4:a 5: a,c

**1.**Question 1

Fill in the blank: A _____ is a collection of data used to teach AI.

training set

computer output

hallucination

brainstorming assistant

1 point

**2.**Question 2

A web designer is using generative AI to create a new website. Which of the following tasks can be accomplished using generative AI? Select three answers.
- [ ] Generate images for the website
- [ ] Predict whether a user will click on a particular link
- [ ] Edit website content
- [ ] Create a variety of website layouts
1 point

**3.**Question 3

A workforce management company is considering the potential challenges of using AI to provide human resources services to its employees. Which of the following issues should they be aware of? Select two answers.
- [ ] The need to continuously train AI on new employee procedures
- [ ] The difficulty AI has with quickly analyzing data
- [ ] The inability of AI to handle data-related tasks
- [ ] The possibility of AI perpetuating existing biases in its training data
1 point

**4.**Question 4

A recruiting agency uses an AI tool to analyze resumes and recommend candidates based on their competencies that match the job description. A staff recruiter then reviews the AI-generated shortlist and contacts these candidates to arrange interviews. What does this scenario describe?

AI augmentation

Commissioning

Filtering

AI automation

1 point

**5.**Question 5

How can businesses effectively integrate AI into their workflows? Select two answers.
- [ ] Maintain human oversight over the AI's output and decision-making processes.
- [ ] Rely entirely on computer engineers to keep AI solutions updated.
- [ ] Promote cross-team collaboration to ensure the AI aligns with values that benefit people.
- [ ] Only use the latest AI tools.
1 point

Module 2: Maximize productivity with AI tools

Types of generative AI Tools

Text Generator
Image Generator
Audio & Video Generator
Code Generator

Understand how AI tools work

AI tool: AI tool is AI powered software that can automate or assist users with a variety of tasks.

stand-alone AI tools, integrated AI features or custom solutions.

A stand-alone AI tool: A stand-alone AI tool describes AI-powered software that's designed to be used on its own. Stand-alone AI tools can be accessed online or downloaded to a device with little or no technical setup.
An integrated AI feature: An integrated AI feature refers to a built-in enhancement to a particular piece of software.
A custom AI: A custom AI solution is an application that's tailor-made to solve a specific problem.

AI models:An AI model is a computer program trained on a set of data to recognize patterns and perform specific tasks.

Think of the AI model as the engine and the AI tool as the car. The model provides the underlying capabilities while the tools interface assists you in completing tasks. Whether you're using a stand-alone tool, an integrated feature, or a custom solution in your work, remember that an AI tool is powered by an AI model. Next up, we'll explore how to effectively engage with AI tools to tackle a variety of tasks and boost your productivity.

AI models and the training process

AI Tool: An AI tool is AI-powered software that automates or assists users with various tasks, built on top of AI models.
Training Process: The training process involves defining a problem, collecting and preparing data, training the model, evaluating its performance, and deploying it in an AI tool.
AI Agent: An AI agent is an AI-powered tool that autonomously performs tasks with minimal human oversight, following predefined rules.
AI Model: An AI model is a computer program trained on data sets to recognize patterns and perform specific tasks, essential for the functionality of AI tools.

AI tools and AI models

AI tools and AI models are different things, even though they sound similar and are closely related. An AI tool is AI-powered software that can automate or assist users with a variety of tasks. An AI model is a computer program trained on sets of data to recognize patterns and perform specific tasks. All AI tools are built on top of AI models, which are necessary for AI tools to work.

To better understand the relationship between an AI tool and an AI model, think about a car and its engine.

The car, with its user-friendly parts, like a steering wheel and dashboard, represents an AI tool, but it helps you complete tasks instead of drive from one place to another. And just like cars have different parts, AI tools have different parts, as well. One of the most important parts of a car is its engine. Without it, the car wouldn’t work. The same is true for an AI tool, which relies on an AI model to be its “engine,” processing the information that you, the user, provide and allowing the AI tool to function.

But not all engines work the same. Similar to how some cars are better for certain things than others— for example, pickup trucks can haul cargo—AI tools are developed for a wide range of applications. There are AI tools for generating text, images, videos, or even computer code. But regardless of the specific function of the AI tool, it needs an AI model to work.

If the car can also drive itself, navigating from one destination to another without you having to adjust the steering wheel, it would operate similar to how an AI agent operates. An AI agent is an AI-powered tool that can autonomously perform tasks with little human oversight. For example, AI agents can automatically respond to emails, post content on social media, or monitor computer networks. You set the rules for how the AI agents operate, and the AI agents complete tasks following those rules, allowing you to work on other tasks.

Note: Some AI tools incorporate multiple AI models, working together to achieve more and perform a wider range of tasks. Each model within the tool might be specialized for a specific subtask, which contributes to the overall functionality of the AI tool. You'll explore these types of multimodal tools later in this course.

The process of training AI models

AI designers develop AI models through a process called training. Here’s an example of the typical steps a designer might take in this process, in this case for building a model that predicts rainfall:

Define the problem to be solved. AI designers want to predict rain to help people stay dry when commuting to and from work. They consider the AI tool’s features and limitations before identifying an AI solution.
Collect relevant data to train the model. AI designers gather historical data of days when it rained and days when it didn't rain over the past 50 years.
Prepare the data for training. AI designers prepare the data by labeling important features, like outdoor temperature, humidity, and air pressure, and then noting whether it rained. It's also common to separate the data into two distinct sets: a training set to use during the training step and a validation set to test after training is complete.
Train the model. AI designers apply machine learning (ML) programs to the prepared training data. As the ML programs analyze the data, they begin learning how to recognize patterns that indicate the likelihood of rainfall, like the combination of high temperatures, low air pressure, and high humidity.
Evaluate the model. AI designers use the validation set they prepared earlier to assess their model's ability to predict rainfall accurately and reliably. Analyzing a model's performance can uncover potential issues impacting the model, like insufficient or biased training data. If any issues exist, the AI designers can revisit an earlier step in the process and try a different approach. Once the model performs well with its validation set, the process continues to the next step.
Deploy the model. When the AI designers are satisfied with their model's performance, they deploy it in an AI tool—helping people in their city stay dry on their way to work.

Model training is an iterative process. AI designers and engineers can repeat each step as many times as necessary and make adjustments until they create the best model possible.

But the process doesn't stop at deployment. Once users interact with a model in practical situations, the model might be exposed to new challenges. AI designers should continuously monitor and collect feedback on their models, ensuring that they continue to perform reliably and to identify areas for improvement. This iterative process of continual refinement makes AI models precise and versatile, resulting in effective, reliable AI tools.

Transform your work with generative AI

**Prompt: (**To use a generative AI tool, you'll need to guide the tool with a prompt.) A prompt is text input that provides instructions to the AI model on how to generate output.
- Think of prompts as your way of communicating with a generative AI tool.

Generative AI tools for workplace tasks

Integration Strategies: Methods for incorporating AI tools into workflows, including using native features, browser extensions, dedicated applications, and automation platforms.
Text and Content Generative AI: These tools leverage Large Language Models to understand and generate human-like text, aiding in tasks like drafting emails and summarizing research.
Generative AI Tools: Generative AI tools are designed to assist users in creative tasks by generating content, code, or media based on user input, utilizing advanced algorithms and large datasets.
Code-Generative AI: Specialized tools that assist software developers by suggesting code completions and generating functions from natural language descriptions.
Image and Media Generative AI: Tools that create and edit multimedia content using diffusion models, allowing users to generate images and videos from text prompts.

Generative AI tools are designed to be creative and collaborative partners. They interact with users conversationally through natural language, making them incredibly versatile for a wide range of tasks like writing, planning, coding, and designing. Some of these tools are standalone applications, while others are seamlessly integrated into the software you use every day. This reading contains an overview of the primary categories of generative AI, key examples, and information about how to strategically integrate them into your work to accomplish more, faster.

Text and content generative AI tools

These tools are built on Large Language Models (LLMs) that have been trained on vast amounts of text and data. This allows them to understand, summarize, translate, predict, and generate human-like text. They function as versatile assistants for any task involving language, from drafting an email to analyzing complex research papers. Their core strength lies in their ability to process and generate language with nuanced understanding of context, tone, and intent.

Examples include:

Code-generative AI tools

Often described as an "AI pair programmer," these tools are specialized for software development. They are trained on billions of lines of code from public repositories, enabling them to understand programming languages, frameworks, and common coding patterns. They assist developers by suggesting code completions, generating entire functions from natural language descriptions, identifying and fixing bugs, writing unit tests, and explaining complex code blocks. They integrate directly into Integrated Development Environments (IDEs) to provide real-time assistance.

Examples include:

Image- and media-generative AI tools

These tools focus on creating and editing multimedia content, including images, video, and audio. They typically use a technology known as diffusion models, which learn to generate novel content from text descriptions (prompts). Users can create photorealistic images, artistic illustrations, marketing materials, and video clips simply by describing what they want to see. These tools are revolutionizing creative workflows by dramatically reducing the time it takes to visualize ideas and produce high-quality media.

Examples include:

Integrating AI tools into your workflow

The true power of AI is unlocked when it moves from a novelty to a natural part of your daily process. Thoughtful integration can save time, reduce tedious work, and enhance creativity.

Common methods of integration:

Native features: Many applications you already use are adding built-in AI capabilities. This is the most seamless form of integration, as the AI has context on the work you are already doing within the app.
Browser extensions: Some tools can be added to your web browser. This allows the AI to assist you across a wide range of web-based applications, from writing emails in Gmail to drafting posts on social media.
Dedicated applications: Using a standalone tool as a "thinking partner" is a common workflow. You can copy and paste text between your work and the AI tool to brainstorm, summarize, or refine content.
Automation platforms: For more advanced users, tools can be connected across different apps. These sophisticated platforms allow you to create automated workflows between different apps (e.g., automatically summarizing an email with AI and adding it to a to-do list) without writing any code.

A practical approach to getting started:

Identify bottlenecks: Pinpoint the most time-consuming or repetitive tasks in your day. Is it writing first drafts? Summarizing meeting notes? Finding bugs in code?
Start small: Choose one bottleneck and find a single AI tool that addresses it. Focus on mastering that tool for that specific task.
Build a habit: Make a conscious effort to use your chosen tool whenever you perform that task. The goal is to make it a natural reflex.
Evaluate and expand: After a week or two, assess the impact. Are you saving time? Is the quality of your work improving? If the integration is successful, look for the next bottleneck and explore how AI can help you there.

A final thought: This list is just a starting point. The world of generative AI is expanding at an incredible pace, with new and more powerful tools emerging continuously. The key is not to become an expert with every tool, but to develop a mindset of curiosity. Take the time to explore tools that suit your specific needs, like Gemini or NotebookLM, and consider how they can augment your skills and support the work of you and those around you.

Activity: Use generative AI to develop content and images for social media

AI tool access

Option 1: Develop a social media post of your own

Think about something you want to share on social media. It could be about work or one of your hobbies. For example, maybe you received an award or finished a novel, but you’re not sure how to share this information in a way that will catch people’s attention. AI tools can help with that.

Option 2: Develop a social media post for a pet care company

Your job is to promote online engagement for a pet care company by sharing information about dog breeds on social media. Your first post needs to highlight fun facts about Golden Retrievers, including a catchy image. You need help getting started and you have a limited amount of time, so you plan to use an AI tool to help.

Let’s get started.

Step 1: Generate text for your social media post

Option 1: Develop a social media post of your own

Prompt the AI tool to draft a social media post. Remember that prompts can take a variety of different forms, and you can phrase a prompt in any way that feels natural to you. Just make sure you provide clear, specific instructions and include any relevant details about the content, as well as any other considerations, like length or tone.

Enter additional prompts to refine the social media post. You might prompt the AI tool to make adjustments based on:

Accuracy: Is all the information in the post correct?
Length: Does the post follow the right character or word count requirements?
Tone: Is the post's tone consistent with the desired style?

Continue to prompt the AI tool to refine your social media post until you are satisfied.

Option 2: Develop a social media post for a pet care company

Enter this prompt into the AI tool: Craft a compelling social media post for dog owners, highlighting a fun fact about Golden Retrievers. The post should be informative and catchy, and include a description of a relevant image that captures the breed's personality. Keep the post under 260 characters. It should promote online engagement by encouraging sharing and comments.

Review the AI tool’s output, keeping in mind the purpose of this task. Consider questions like:

What fun fact did the AI tool come up with?
Would the social media post appeal to pet care company's customers?
What aspects of the social media post do you want to change?

Enter additional prompts to refine the social media post. You might prompt the AI tool to make adjustments based on:

Accuracy: Is all the information in the post correct?
Length: Does the post follow the right character or word count requirements?
Tone: Is the post's tone consistent with the desired style?

Continue to prompt the AI tool to refine your social media post until you are satisfied with the output.

Step 2 : Generate an image for your social media post

Option 1: Develop a social media post of your own

Prompt the AI tool to generate a custom image that you can include in your social media post. Make sure to provide vivid descriptions. For example, you might specify the size, color, and position of subjects in the image, or the style you want.

Review the AI tool’s output, and evaluate how well it aligns with the purpose of this task. You might prompt the AI tool to make adjustments based on:

Composition: Is the image well-structured and visually appealing? Are the elements arranged effectively?
Clarity: Is the image clear and easily understandable? Does it accurately represent the described scene or subject?
Style: Does the image match the desired artistic style (e.g., realistic, abstract, cartoonish)?

Continue to prompt the AI tool to refine the image until you are satisfied with the output.

Option 2: Develop a social media post for a pet care company

Enter this prompt into the AI tool: Create a bright and warm image of a golden retriever eating kibble. The dog should be looking directly at the viewer with a friendly expression. Use a cartoonish style with a bright, cheerful background. The image should be suitable for a social media post promoting a pet care company.

Review the AI tool’s output, and evaluate how well it aligns with the purpose of this task. You might prompt the AI tool to make adjustments based on:

Composition: Is the image well-structured and visually appealing? Are the elements arranged effectively?
Clarity: Is the image clear and easily understandable? Does it accurately represent the described scene or subject?
Style: Does the image match the desired artistic style (e.g., realistic, abstract, cartoonish)?

Continue to prompt the AI tool to refine the image until you are satisfied. If you post the image on social media, a responsible practice is to disclose that you’ve used an AI tool to create the image. (You’ll learn more about using AI responsibly in Module 4.)

Leverage the human-in-the-loop approach to AI

Human-in-the-loop approach: a combination of machine and human intelligence to train, use, verify, and refine AI models.
- A human-in-the-loop approach blends the efficiency of AI tools with human insight that's critical for practicing responsible AI.
Responsible AI: Responsible AI is the principle of developing and using AI ethically with the intent of benefiting people and society while avoiding harm.
- One aspect of responsible AI involves managing the limitations of AI models such as knowledge cutoff.
Knowledge cutoff: Knowledge cutoff is the concept that an AI model is trained at a specific point in time so it doesn't have any knowledge of events or information after that date.

!!! There are situations when a tool might try to generate a response despite its knowledge cutoff, leading to a hallucination.

Hallucination: Hallucinations are AI outputs that are not true.
- Hallucinations can be problematic because they can lead to misinformation, misinterpretation, or inappropriate responses that might damage a company's reputation or result in customer dissatisfaction.
- For example, picture working at a retail company using an AI tool to forecast how much product inventory to order. If the AI tool produces inaccurate, outdated, or misinterpreted information, those hallucinations could lead the company to order the wrong amount of inventory causing supply issues for customers.
- While hallucinations can pose challenges, they can also be beneficial to your creative process. Suppose you use an AI image generator to help you design concept art for a fantasy-themed video game. You prompt the AI tool to create an image of a beautiful castle floating in the sky, and it outputs a unique, fun image, which also happens to be a hallucination. In this example, you produced a hallucination intentionally.

By reviewing and evaluating AI-generated content, you can help mitigate the potential effects of hallucinations. This approach helps ensure that AI-generated outputs are not only innovative, but also accurate, relevant, and ethical, enhancing outcomes for businesses, customers, and society as a whole.

CodeAI: CodeAI is ways to generate programming languages in code so that software developers can use CodeAI to help them write applications.

Determine if generative AI is right for the task

Assess whether to apply generative AI to a task:
- First, is the task generative?
- Can the task be iterated on to achieve the best outcome?
- Are there resources to provide adequate human oversight?

Use Gemini in Google Docs, Slides, Sheets, Meet, and Gmail

Gemini: Gemini is an AI tool available in Google Workspace apps that assists users in various tasks such as writing, summarizing, and visualizing information.
Google Slides: A presentation tool where Gemini can generate images and apply visual styles.
Google Sheets: A spreadsheet application where Gemini helps in data analysis and visualization.
Gmail: An email application where Gemini assists in drafting and editing emails.
Google Meet: A video conferencing tool where Gemini can generate captions and take notes during meetings.
Google Docs: A collaborative application for creating documents where Gemini can help with writing, summarizing, and proofreading.

Discovering new AI tools is a helpful way to stay up to date with emerging technology and boost your productivity. In this reading, you'll learn about Gemini in Workspace apps, including Gmail, Google Docs, Slides, Sheets, and Meet. You'll also explore examples of how you can prompt Gemini in Workspace apps to help you write, visualize, organize information or projects, and better connect with others.

Introduction to Gemini in Workspace apps

You can use Gemini in Gmail, Google Docs, Slides, Sheets, and Meet. It’s available as an add-on for purchase for existing Google Workspace accounts. You can access Gemini in all of these Workspace apps on desktop computers. You can access Gemini in Gmail on both desktop computers and mobile devices.

Note: Even if you’re not currently using an eligible Google Workspace account, you may be able to access Gemini in Workspace apps in the next activity. Gemini in Workspace apps is not yet available in certain countries and languages. For more details, refer to documentation about Where you can use Google Workspace Labs.

On the Google Workspace website, you can review information about Gemini in Workspace apps, including visual examples of its functionality and information about purchasing eligibility. Consider exploring Gemini in Workspace's Prompting guide 101 for more prompting guidance.

Google Docs

Google Docs is a collaborative application for creating many types of documents. You can use Gemini in Docs to perform a variety of tasks related to writing and refining documents, such as to:

Write text for your document
Summarize the content of your document
Brainstorm new ideas to include in your document
Suggest stylistic changes to existing text
Proofread for grammar and spelling

For example, a real estate agent might use Gemini in Docs to help write a new blog post about a nearby town with homes for sale. They can prompt Gemini in Docs for suggestions if they have trouble getting started, or to brainstorm specific ideas. After the business owner reviews and customizes the output, they can add another prompt to proofread the post.

Gmail

Gmail is an application for sending and receiving emails. Similar to Gemini in Docs, Gemini in Gmail can help you with writing-focused tasks. You can prompt it to:

Draft emails
Edit written content by formalizing, shortening, or elaborating on existing text

For example, an account manager wants to send a department-wide invitation email for the quarterly review business meeting. The manager can prompt Gemini in Gmail with a short description of the email’s purpose, such as an invitation to the quarterly review business meeting. Then the manager can insert the suggested text into the body of the email. The manager can also use the Formalize option to make the tone of the email more formal. Lastly, they can check, edit, and refine the text to better meet their needs, as well as fill in details like the date and location of the meeting.

Google Slides

Google Slides is a collaborative application for creating slide-based presentations. When working with Gemini in Slides, you can:

Generate unique images to convey your ideas visually
Apply specific visual styles to generated images

The owner of a small coffee shop, for example, might use Gemini in Slides to create a marketing campaign for the upcoming launch of a holiday coffee blend. The business owner can request a specific style for their images, such as a photograph with a solid background, a sketch, a watercolor painting, and more. Then, the business owner can iterate on their prompt until they generate an image in a style they want to use in their upcoming holiday ad campaign.

Google Sheets

Google Sheets is a collaborative application that lets you organize and analyze data in spreadsheets. Gemini in Sheets can help you to:

Build trackers for a project
Analyze data and get insights
Generate visualizations, like charts or graphs

For instance, a product manager might use Gemini in Sheets to help them make sense of feedback about a product. They could ask Gemini in Sheets to analyze survey data, offer recommendations, or create a visualization.

Google Meet

Google Meet is a video-conferencing application for face-to-face virtual meetings. You can use Gemini in Meet to:

Generate captions translated to and from certain languages during meetings
Create unique background images
Take notes for you

For example, consider a salesperson who wants to take notes during a meeting with a client so they can share the notes with their team members. The salesperson doesn’t want to be distracted from the meeting, so they decide to use Gemini to take notes, allowing them to stay focused on their business with the client while creating a thorough record to share afterward.

Activity: Increase productivity with the help of Gemini for Google Workspace

As a quick reminder, here's how generative AI tools work:

You provide input (i.e., a prompt) to the AI tool.
The AI tool processes the input.
The AI tool produces output.

Remember that providing human oversight is crucial when using an AI tool, so always be sure to review AI-generated output and give additional instructions to the tool as needed.

AI tool access

To access Gemini in Workspace sign up for Workspace Labs with a Google Account. Gemini in Workspace apps is not available in certain countries and languages. For more details, refer to Where you can use Google Workspace Labs.

To access Gemini in Workspace apps:

Sign in to your Google Account.
Sign up for Workspace Labs.
After user sign-up, Workspace Labs is turned on. An in-product welcome screen will appear the next time you open a Workspace app, such as Google Docs and Sheets.

Refer to the resource about how to Create a Google Account, if you don't already have one. For more information on Workspace Labs, you can also review the documentation on how to Get started with Google Workspace Labs.

Note: Before you use Gemini in Workspace apps, review the following information:

Workspace Labs is a program for testing Google Workspace features before they become broadly available.
For more details on using Gemini, such as who can use Gemini, Gemini’s Privacy Notice, and where Gemini is currently available, refer to the Gemini Apps FAQ.
Review the Google Workspace Labs Privacy Notice and Terms for Personal Accounts.
Please don’t enter private or confidential information in your Gemini in Workspace apps conversations or any data you wouldn’t want Google to use to improve its products, services, and machine learning technologies. Google uses Workspace Labs Data and metrics to provide, improve, and develop products, services, and machine learning technologies across Google, including Google’s enterprise products.
While Gemini in Workspace apps is improving every day, users may see slightly different experiences in their Workspace instance as feedback and new features are incorporated into the product.
Feedback from a wide range of experts and users helps Gemini in Workspace apps improve every day. You can provide feedback using the thumbs up or thumbs down button—with the option to further explain in a comment.

Read through the following options. Then complete the activity using the option that best suits you.

Option 1: Explore Gemini in Workspace apps in your daily life

Choose a task or project in your professional or personal life that involves any of the activities listed below. For example, you might need to plan an event and communicate details to attendees, or you might need to create a budget and share it with someone. Just make sure the task involves at least one of these activities:

Drafting a document
Developing a presentation
Working in a spreadsheet
Sending an email
Taking notes during a virtual meeting

Option 2: Manage a bookstore with help from Gemini in Workspace apps

You’re the manager of an independent bookstore, and you want to understand the profits from your fiction book sales. You decide to use Gemini in Google Workspace apps to help you with this task by creating:

A document reporting information about sales
A presentation to summarize the information about sales to your staff
A spreadsheet to track sales
An email to invite suppliers to a presentation you’ll give about book sales
Taking notes during a virtual meeting

Generate insights from multiple sources with NotebookLM

Citations: NotebookLM includes citations in its responses, allowing users to verify the accuracy of the AI's summaries by referencing the original source material.
Source Grounding: NotebookLM's responses are based only on the sources provided by the user, ensuring that the AI's answers are predictable and verifiable.
User Interface Panels: The NotebookLM interface consists of three main panels: Sources for managing materials, Chat for interacting with the sources, and Studio for transforming sources into new formats.

Have you ever wished you could have a conversation with your documents? Imagine asking questions of a dense report or getting a quick summary of scattered notes. That’s the idea behind NotebookLM, a research assistant and thinking partner that uses generative AI to help you explore your own materials. What separates it from other AI tools is that its answers only rely on the information from sources you provide. In this reading, you’ll explore its main features using a public notebook of Shakespeare’s complete works.

An AI tool with source grounding

The most important feature of NotebookLM is that its responses are grounded only in the sources you provide. This means that when you enter a prompt, NotebookLM searches your uploaded sources to construct an answer and provides clear citations to show you exactly where the information came from. Think of it like an open-book test where the AI is only allowed to use the documents you've given it.

This has some key benefits:

It makes the tool more predictable. You control the information the AI references based on what resources you choose to include. This makes the AI's behavior predictable and keeps the focus right where you want it.
Fact-checking is simple. NotebookLM’s answers include citations. You can check if the AI's summary is accurate by revisiting the source material, which helps build trust in the tool’s output.

Explore NotebookLM with Shakespeare

Now it's time to explore NotebookLM in action. To get familiar with the tool, we are going to use a Notebook that is pre-loaded with the complete plays of William Shakespeare. Inside, you'll find the notebook consists of three main panels:

Sources: This is where you add and manage the materials of a notebook.
Chat: This is your primary workspace for interacting with your sources by prompting it with your questions and ideas.
Studio: This is where you can transform your sources into new formats like study guides, quizzes, mind maps, videos, and podcasts.

First, get familiar with the Source panel. Click on a source, like "A Midsummer Night's Dream." Notice how this displays the full text of the play and an AI-generated overview at the top, giving you a quick summary of the material.

Next, use the Chat panel to prompt NotebookLM about the texts. For example, you could enter a prompt like:

In Romeo and Juliet, what are the main events that lead to the tragic ending?

Notice how the AI-generated response includes citations. Clicking these will show you the exact passages used by the tool to generate its answer. Now, try a prompt on your own! You could ask for a character analysis, a summary of a specific act, or a comparison of two plays.

Finally, explore the Studio panel for examples of how source materials can be transformed into new formats. Check out a few of the artifacts that have been pre-generated for you, like:

An audio overview of Shakespeare's tragedies
A video overview of Hamlet
A study guide of Macbeth
A mind map of Shakespearean themes and characters

Now that you’ve gotten a sense of NotebookLM in action, you’re ready to test it out with your own sources! Get started at notebooklm.google.com.

Wrap-up

NotebookLM is a specialized tool designed to help you analyze, understand, and find new connections within the materials you provide. Unlike a general-purpose AI tool, its responses are grounded only in the sources you select. Whether you're working with project reports, interview transcripts, or research articles, you can use NotebookLM to turn a collection of documents into an interactive knowledge base—helping you work smarter with the help of AI.

Resources for more information

Check out these resources to learn more about how you can make NotebookLM work for you:

Module 2 Challenge (Quiz)

1.C | 2.D | 3.A | 4.B | 5. B,C,D

1. Which of the following statements best describes an AI tool?

An AI tool is a hardware component designed to boost the performance of AI models.

An AI tool is a type of machine learning model designed specifically for advertising campaigns.

An AI tool is AI-powered software that can automate or assist users with a variety of tasks.

An AI tool is the process of using AI to accomplish tasks without any action on the user’s part.

2. A teacher uses a generative AI tool to brainstorm ideas for a new lesson plan. They provide the tool with text input describing the content of the lesson and how the lesson should be formatted. What is the term for this text input?

A submission

A query

A direction

A prompt

**3.**A fashion designer is developing a new fall collection. They need to create detailed illustrations of the clothes before their manufacturer begins production. What type of generative AI tool can help them with this task?

A code generator

An image generator

An audio generator

A text generator

**4.**What is the main purpose of applying a human-in-the-loop approach when using AI?

Increase the processing speed of AI models

Ensure that AI outputs are useful and safe

Reduce the need for human oversight

Enhance AI's ability to learn independently

**5.**You are considering whether to use a generative AI tool to help you with a task at work. Before proceeding, which guiding questions should you answer “yes” to? Select three answers.

[ ] Is the task critical to the final product?
[ ] Is the task generative?
[ ] Can the task be iterated on to achieve the best outcome?
[ ] Are there resources to provide adequate human oversight?

Module 3: Discover the Art of Prompting

prompt: A prompt is text input that provides instructions to the AI model on how to generate output.
- The more clear and specific your prompt, the more likely you are to get useful output. Another important part of prompt engineering is iteration. You'll learn about evaluating output and revising your prompts. This will also help you get the results you need when leveraging conversational AI tools in the workplace.
A large language model, or LLM: A large language model, or LLM is an AI model that is trained on large amounts of text to identify patterns between words, concepts, and phrases, so that it can generate responses to prompts.
- An LLM is trained on millions of sources of text, including books, articles, websites, and more. This training helps the model learn the patterns and relationships that exist in human language. In general, the more high quality data the model receives, the better its performance will be.
- LLMs use statistics to analyze the relationships between all the words in a given sequence and compute the probabilities for thousands of possible words to come next in that sequence. This predictive power enables LLMs to respond to questions and requests, whether the prompt is to complete a simple sentence or to develop a compelling story for a new product launch or ad campaign.
- Although LLMs are powerful, you may not always get the output you want. Sometimes this is because of limitations in an LLM's training data. For instance, an LLMs output may be biased because the data it was trained on contains bias.
- Another factor that can affect output is the tendency of LLMs to hallucinate.
Hallucinations: Hallucinations are AI outputs that are not true.
- While LLMs are good at responding to many kinds of questions and instructions, they can sometimes generate text that is factually inaccurate.
- Because of an LLM's limitations, it's important that you critically evaluate all LLM output to determine if it is factually accurate, is unbiased, is relevant to your specific request, and provides sufficient information.
Prompt engineering involves designing the best prompt you can to get the output you want from an LLM. This includes writing clear, specific prompts that provide relevant context.

Prompting best practices

References: Examples or resources that illustrate the desired output style, tone, and format for the AI tool.
Iterate: The practice of refining prompts based on AI output to achieve better results through continuous improvement.
Task: The specific action or output you want the AI tool to perform, clearly stated to avoid ambiguity.
Context: Detailed background information and objectives that help narrow the AI tool's focus and tailor its output.
Evaluate: The process of assessing the quality and effectiveness of AI-generated content before use.

Prompting is an essential skill for getting the most out of generative AI. To get the best results from an AI tool, you need to craft a good prompt. A good prompt follows a simple framework: Task, Context, References, Evaluate, and Iterate. Explore how you can use this framework to elevate your prompting and unlock more of AI’s benefits.

When you provide an AI tool with a task, also consider including information about:

Persona: What expertise do you want the AI tool to draw from, and who is the audience? For example, you might ask the tool to complete the task with the expertise of an IT professional, or ask it to create an output geared towards a specific audience like your manager or team.
Format: How do you want the output to be formatted? For example, you might ask the AI tool to create a bulleted list or a comparison table.

Those additional pieces of information can help the AI tool accomplish the task the way you want it done.

Here's an example of a prompt that includes a specific task, a persona, and a format:

Task: Create a social media post about an upcoming music festival that speaks to the local music community while attracting out-of-state festival-goers.

Persona: You're a concert promoter specializing in raising ticket sales in the alternative rock music industry.

Format: Limit the post to 125-characters. Include 5 relevant hashtags.

Provide necessary context

Including detailed context in your prompts can narrow an AI tool's focus and tailor its output. Contextual information in your prompts might include:

Reasons and objectives for performing the task
Rules or guidelines that the output must follow
Relevant background information the tool should consider

Those details can help an AI tool better understand your needs. Here's an example of a prompt that provides necessary context:

Include references as examples

References are examples or resources that illustrate what you want an AI tool to produce. They specify details about your desired output, such as the style, tone, and format. Depending on the AI tool, you can include text, images, audio, or even video as references.

When including references in your prompts, consider these suggestions:

Briefly explain how the references relate to the task.
Use 2-5 high-quality examples that closely align with your needs.
Include your own work or open-source examples if relevant.

Evaluate your output

AI tools vary in their training and capabilities. Each tool has unique strengths and limitations, which can influence the quality of their output. After receiving a response from an AI tool, it's essential to carefully evaluate the quality and effectiveness of the AI-generated content before using it or sharing the output with others.

When evaluating the output, focus on factors such as:

Accuracy
Bias
Relevancy
Consistency

AI-generated content should serve as a starting point, not a final product. Sometimes while assessing and validating an AI output, you might determine that it's unacceptable or not useful. When that's the case, you should continue on to the next step of the prompting framework.

Iterate for better results

Even well-crafted prompts might not produce ideal results on the first try. This is where iteration comes in—the process of refining your prompt based on the AI's output. Think of iteration like having a back-and-forth conversation with an AI tool. The process typically goes like this:

You provide an initial prompt.
The AI tool responds with an output.
You evaluate the effectiveness of the AI-generated response.
You refine your request based on what worked and what didn't.
The process repeats until the AI tool produces the desired results.

Effective prompting is not about getting a perfect result on the first try, but about being willing to continuously improve your approach. Be patient, provide clear feedback, and keep prompting until you reach the desired outcome.

Pro Tip: When you've arrived at an effective prompt for a particular task, save it! You can use your most effective prompts as templates for different use cases and needs. This can help you replicate successful AI-output consistently—without starting from scratch every time.

Activity: Write an effective prompt for Gemini

Scenarios

Option 1: Write a prompt to delegate work

Think of a piece of work you want to delegate to a new team member at work. It could be something like drafting a document or doing background research. For example, you might need to create a template for earnings reports that your team submits. AI tools can help with that.

Option 2: Write a prompt to generate a marketing email for a new business

You are a florist who's opening a new business that specializes in bespoke arrangements at large events. You think other businesses in the area might be interested in your services, and you want to market your business to them via email. You think Gemini could help.

Let's get started.

Step 1: Write a prompt

For option 1

If you were to delegate the piece of work to a teammate, what would you tell them to set them up for success? You’d probably describe the task in detail, give lots of background information, and provide an example or two. Clear and specific guidance would help your teammate meet your expectations and do great work.

You can work with Gemini the same way – by giving it clear and specific instructions, or prompts. To get the best results from your prompts, follow the TCREI framework: describe your task, specify the context, and give plenty of references.

Using this framework, draft a prompt, then enter it into Gemini.

For option 2

Work through the first 3 steps of the TCREI framework to make sure your prompt is effective.

Tell Gemini the task you want it to complete. Use clear and direct language. You might even include a persona and specify a format. In this case, the task might be to write an introductory email about the new floral design business to other local businesses, with the goal of exploring potential partnerships*.* The persona could be an email marketer who specializes in new business development, and the format could be 500 words with a friendly tone, bullet points, and an engaging subject line.

Give Gemini any context it needs to help you with the task. The context could include your goals, the reason for the task, or even what you've tried before that didn't work. In general, the more context you provide, the better the output. In this case, the context could include information about how you source flowers locally, specialize in bespoke arrangements, and offer custom pricing for loyal customers.

Provide references, examples you give to Gemini to learn from or emulate while generating output. You can paste a reference directly into your prompt, enter a link from the web, or attach a file.

Enter your prompt into Gemini.

Step 2: Evaluate your prompt

Once you've entered your prompt and received output, you need to evaluate the output. Ask yourself if Gemini's output is accurate, unbiased, and useful. This is part of being a responsible AI user.

Pro tip: Save your most successful prompt in your prompt library to update and reuse for similar tasks.

Step 3: Keep iterating

If you've evaluated the content and you still don't have the output you want, keep iterating. Modify the task, context, or references, or try out an entirely different prompt.

Prompts for different purposes

Extraction: The process of gathering specific information from complex documents using clear prompts.
Prompting Framework: A structured approach to crafting prompts for AI tools, consisting of Task, Context, References, Evaluate, and Iterate.
Translation: Facilitating communication by translating content for global audiences while maintaining tone and structure.
Problem-solving: Using AI tools to break down complex challenges into manageable steps to generate potential solutions.
Content Creation: Using AI tools to generate various types of content, such as text, images, and videos, by providing specific prompts.
Summarization: The ability of AI tools to condense large amounts of text into concise summaries while preserving key information.
Editing: Refining existing text to ensure clarity and effectiveness in communication through specific editing instructions.

Effective prompting can help guide an AI tool to generate useful output for a variety of tasks, from content creation to problem solving. In this reading, we provide some examples demonstrating how to use the prompting framework—Task, Context, References, Evaluate, Iterate—to get reliable and useful results in different cases.

Content creation

AI tools can generate a wide range of content, no matter your industry or role. You can use AI tools to help you create content like images, videos, documentation, and code. When prompted effectively, AI tools can help you spark creative ideas while saving time.

Here's a sample prompt that incorporates prompting best practices by including a specific task, context, and references:

Act like you are a creative advertising professional who can apply innovative thinking to develop original taglines that highlight the positive qualities of a product. Create a concise tagline for a washing machine that gets clothes extra clean, has 25 settings, and fits in a small space. The tagline should use an active voice, and be no more than 6 words. Refer to these examples of successful taglines in this industry, which emphasize efficiency and convenience.

Reference #1: "Less soap. More clean."
Reference #2: "Transform laundry into pure magic."

This prompt clearly states that the task is to create an original and concise tagline. It also provides necessary context by specifying the tone, style, and format of the tagline. The references include examples to help guide the AI tool's output.

Pro tip: Learn more about how you can use Gemini as a creative partner.

Summarization

AI tools can quickly process and condense large amounts of text while preserving key pieces of information—making them ideal for distilling complex information into clear, concise highlights.

Here's a sample prompt that incorporates task specifics, context, and format preferences:

Summarize the following email from a software vendor. Create a concise, professional overview that includes a bulleted list of each paid subscription tier with its benefits to clients. The summary should be easy for a sales representative to quickly review before following up with the potential client.

[Email content would follow here]

This prompt demonstrates how specifying your style, tone, and format preferences can help generate summaries tailored to a specific audience's needs—ensuring the output is useful and relevant.

Pro tip: Learn more about how you can use Gemini to summarize meetings.

Classification

AI tools can help you analyze and categorize content like service emails and customer reviews based on specific criteria. This feature can help you sort through information and organize it into meaningful and actionable insights.

Here's a sample prompt that specifies a task, provides clear classification criteria, and includes references:

Analyze these customer reviews and tell me whether the sentiment of the reviews is positive, negative, or neutral. For each review, provide a one-sentence explanation for the classification based on these criteria:

Positive: Consistently favorable comments or mostly positive with minor criticisms
Neutral: Mixed or balanced positive and negative comments
Negative: Consistently unfavorable comments or mostly negative with minor positive elements
Review #1: I don't know where to begin. We had reservations for 7:00 but they seated us at 7:45. Then, no one came to our table for at least 30 minutes. Our appetizer and main course were mediocre. I did love the dessert, but that wasn't enough to change our experience.
Review #2: I love this restaurant. The food is delicious and the service is excellent.

This prompt begins by clearly stating that the task is to analyze the sentiment of a customer review and then specifies the options: positive, negative, or neutral. By clearly defining each category, the AI tool can make more accurate assessments and generate more consistent results.

Extraction

AI tools can save you time by quickly gathering exactly the information you need from large, complex documents. This could be numbers from financial reports, product details from catalogs, or key points from meeting notes. To achieve more precise results, you can use the verb "extract" in your prompt, along with clear instructions about the type of information you need and how you want it formatted.

Here's a sample prompt that incorporates task specifics, relevant context, and format preferences:

I'm planning our department's clothing budget and need to track pricing trends. Extract all of the references to items of clothing I can buy and how much each item costs from the blog post below. Create a spreadsheet of just these items. Format the output with two columns labeled "Item" and "Price." Capitalize list items and organize them by price from high to low.

[Blog post content would follow here]

This prompt begins by clearly stating that the task is to extract all the items of clothing mentioned in the blog with their corresponding prices. It also specifies that the data should be organized in a spreadsheet, with clear formatting instructions—preparing the data for quicker analysis.

Translation

AI tools can help facilitate communication by translating content for global audiences, making it easier to communicate with international colleagues and create multilingual content.

Here's a sample prompt that incorporates task specifics, context, and style preferences:

We're a boutique bike shop that sells cycling and outdoor adventure gear. Translate our product descriptions from English to Dutch. Maintain the same structure and casual tone that is used in the English version in the Dutch translation. Consider cultural context and ensure the expressions maintain their marketing appeal in Dutch.

Reference #1: Whether you’re exploring city streets or forest paths, our sleek and durable bicycle has it all.
Reference #2: Roll into summer in style with our smooth and stylish rollerblades.

This prompt begins by clearly stating that the task is to translate product descriptions from English to Dutch. It also specifies that the Dutch translations should maintain a similar structure and tone as the English originals, ensuring the tool generates output that resonates with the target audience while staying true to the original content's intent.

Editing

AI tools can help refine and improve existing text to ensure clear communication. Using specific editing instructions helps AI tools make appropriate adjustments while preserving the original message.

Here's a sample prompt that incorporates task specifics, context, and audience needs:

As the communications manager for a manufacturing company, I need to explain our site selection process to local community stakeholders. Rewrite this text for a non-technical audience. Simplify the language while keeping all the main ideas intact.

[Text content would follow here]

This prompt demonstrates how specifying your task and the target audience can lead to more useful revisions when editing text. By clearly stating what needs to be simplified, the AI tool can make appropriate changes while maintaining the core message.

Problem-solving

AI tools can also help you break down complex challenges and quickly come up with potential solutions. When using AI for these kinds of tasks, breaking your request into manageable steps can help to produce better results.

Here's a sample prompt that demonstrates how to structure a complex problem:

We are running a community program to teach children gardening skills. The program runs from June 1 to August 15. We want the children to be able to grow plants that will be ready for harvest by the time the program ends. First, identify a list of 10 plants that can be planted and grown in that time period. Include sources that support the time to harvest for each plant.

Next, choose three plants from the list that will provide the children with the most variety. We want the children to grow three plants that are as different from each other as possible.

This prompt begins with useful context about the program, such as its main purpose and timeline. It also breaks down what would otherwise be a complex task into manageable steps. As with any other use case, applying effective prompting techniques and adjusting them to fit your specific needs is the key to getting reliable and useful results from AI tools.

The examples in this reading demonstrate how your prompting approach should be adapted for different workplace tasks. While each use case can require specific techniques, the core principles remain the same—Task, Context, References, Evaluate, Iterate. With practice, you can become more skilled at crafting prompts for any situation and get the exact output you need.

Activity: Iterate on prompts to find solutions

Scenarios

Read through the following options, then complete the activity using the option that best suits you.

Option 1: Solve a workplace problem of your own

Think about a problem you might encounter in the workplace. Perhaps it's a routine task that you find time-consuming, or a unique challenge you need to brainstorm solutions for. For example, maybe you have lengthy meeting notes that you need to quickly summarize into a condensed bulleted list, or maybe you need help prioritizing a to-do list according to importance and difficulty.

Option 2: Solve a scheduling dilemma problem

As part of an education outreach program, you are organizing a computer literacy course that meets two hours a week. Your nonprofit organization has a limited number of computers. There are only 20 computers available for 35 registered students. You decide to use an AI tool to find solutions to this challenge.

Step 1: Prompt to generate solutions

Whichever option you’ve chosen, prompt the AI tool to generate solutions.

The prompt should begin with a clear description of what task you want the AI tool to do. The prompt should also include important context, such as background information that can help the AI tool understand the situation, like key objectives, relevant requirements, the target audience, and any cost or time considerations.

Once you've crafted your prompt, enter it into the AI tool. In the next step, you'll evaluate how well it worked.

Step 2: Evaluate the output

Review the AI tool's response carefully.

Identify areas where the output could be more helpful, specific, or better aligned with your needs. Ask yourself:

Are the suggestions practical and actionable?
Did the AI tool make any incorrect assumptions, misinterpret the context, or omit any information?
What important factors weren't addressed?

Consider anything that's missing or that could be improved.

Step 3: Iterate for better results

Based on your evaluation of the AI tool’s output, iterate on your prompt until you get the exact output you need. This might require you to provide more details or ask specific questions.

For Option 1:

Consider these questions as you iterate on your prompt:

Did you clearly specify who the solution is for and why it matters?
Could adding a preferred format make the response more useful?
Would including your end goal help generate more relevant solutions?

For Option 2:

Incorporate these additional details about the scheduling dilemma:

The course runs for 8 weeks.
Some students are complete beginners while others have basic computer skills.
The facility is available Monday through Friday, 9 AM-5 PM.

Continue evaluating and iterating until you’re satisfied with the solution.

You should critically evaluate all LLM output by asking yourself the following questions.
- Is the output accurate?
- Is the output unbiased?
- Does the output include sufficient information?
- Is the output relevant to my project or task?
- And finally, is the output consistent if I use the same prompt multiple times?
- If you identify any issues when you evaluate output, iterating on your initial prompt can often help you resolve these issues and get better output.
- Your choice of words can also significantly impact an LLM's output. Using different words or phrasing in your prompts often yields different responses from the model.

Shot: In prompt engineering, the word "shot" is often used as a synonym for the word "example." There are different names for prompting techniques based on the number of examples given to the LLM.
- Zero-shot prompting is a technique that provides no examples in a prompt
- One-shot prompting provides one example
- Few-shot prompting is a technique that provides two or more examples in a prompt.

Techniques for mastering complex tasks with AI

Chain-of-thought prompting: A technique that requests an LLM to explain its reasoning process step-by-step, enhancing transparency and structure in its responses.
Iterative flow: The continuous use of outputs from one prompt as inputs for the next, creating a seamless progression through tasks.
Prompt chaining: A method that breaks down complex tasks into a series of smaller, connected prompts, allowing for a more organized and efficient workflow.
Task analysis: The process of breaking down a complex task into smaller, logical steps to facilitate easier management and execution.
Checkpoints: Strategic points in the prompting process where the AI is asked to summarize the overall goal, ensuring it stays on track.

Breaking down complex tasks into manageable steps can be a challenge when working with large language models (LLMs). Fortunately, specialized techniques can help you guide the AI through more complicated work. Two powerful approaches are chain-of-thought prompting and prompt chaining.

Chain-of-thought prompting

Chain-of-thought prompting is a prompting technique that involves requesting an LLM to explain its reasoning process step-by-step, from input to final output. This technique essentially asks the AI to "show its work," making its response more transparent and structured.

To use chain-of-thought prompting, simply include key phrases in your prompt, like:

“Explain your reasoning”
“Go step by step”

These additions indicate that the AI needs to trace its thought process, which often leads to more informative and accurate output.

Let’s go through an example when this technique can be useful:

Consider an HR manager who is developing onboarding materials for a specific department. Here’s how chain-of-thought prompting could be used to identify actionable steps to handle the task:

Prompt:

Create a bulleted list outlining the major duties and responsibilities of a new entry-level design hire at an ad agency. Explain your reasoning step by step.

Notice the additional instructions “explain your reasoning step by step”? You can try using this prompt with and without this phrase to see the difference in AI’s response.

By asking the AI to break down the logic and reasoning behind the duties it suggests, the information you get from the output will also have a rationale for why it was suggested. In this case, it may help the HR manager understand each identify potential gaps and make informed decisions about how to improve their current onboarding process.

Prompt chaining

While chain-of-thought prompting focuses on the reasoning within a single prompt, prompt chaining helps you tackle large projects by breaking them into a series of smaller, connected steps that are all in the same chat. It works like a factory assembly line: the output from one prompt is used as the input for the next, linking all your tasks together like a chain.

This technique involves three key steps:

Task analysis: Start by breaking down your complex task into a series of smaller, logical steps.
Initial prompting: ****Craft a focused prompt that asks the AI to complete just the first step.
Input/output flow: Use the output from the first prompt as the context for the second prompt. Continue this iterative flow until you complete the task.

Let’s take a look at an example of a prompt chain:

Consider that you are planning a short vacation to Paris. You want to make sure you create an itinerary of the things you love most in a vacation, while maximizing logistics to be as efficient as possible with your time:

Prompt 1: ****I'm going to [city name] for 3 days. I like art, historical sites, and parks. Suggest a few well-known places I could visit on my trip.

Prompt 2 (chained from Prompt 1): ****Using those locations, create a logical, day-by-day itinerary that minimizes travel time.

Prompt 3 (chained from Prompt 2): ****For each day of the itinerary, suggest a few restaurants located near each of the suggested locations.

By breaking the task into logical, digestible steps and using the output of one prompt as the specific input for the next, you transform the AI from a simple answer-generator into a structured collaborator.

Combining prompt chaining with chain-of-thought

You can combine prompt chaining with chain-of-thought prompting to enhance the quality and accuracy of the problem-solving process at any stage.

Let's take a look at an example where we combine chain-of-thought prompting and prompt chaining:

Imagine you are hosting your next book club meeting and you want suggestions for what book to recommend. You would like to recommend a fantasy novel, but fantasy is not a genre your book club typically reads. The fantasy book you select needs to be engaging without being overwhelming or too complex.

Prompt 1: I am hosting a book club and would like fantasy book recommendations for people that are new to reading this genre. Suggest a few books that we could use.

Prompt 2 (using prompt chaining and chain-of-thought prompting): From that list, can you suggest which book you would recommend if we are looking for a fast-paced read? Explain your reasoning.

This iterative process continues until a final book is selected and vetted. By breaking down the complex task with prompt chaining and incorporating chain-of-thought, you can address more difficult problems while ensuring more accurate results.

Limitations and best practices

While prompt chaining can be a useful technique, it's important to be aware of potential challenges that can occur—particularly with longer chains.

AI tools can struggle to remember context from earlier parts of the conversation as the prompt chain grows longer. This can lead to:

Inconsistent responses
Overlooking important details from earlier prompts
Difficulty maintaining the overall objective of the task

To overcome these challenges and ensure effective prompt chaining, consider the following strategies:

Use checkpoints: Periodically ask the AI to provide a brief summary of the overall goal.
Work on sub-tasks: Divide very complex tasks into even smaller sub-tasks so that you can treat each as its own shorter chain before moving on to the next.
Recap and redirect: If you notice an AI tool is deviating from the original goal, provide a recap of the essential information and redirect it back to the main objective.

By employing these strategies, you can generate more accurate and relevant output from an AI tool. As with any other prompting technique, prompt chaining relies on finding the right balance between providing necessary context and avoiding information overload.

Module 3 challenge

1: b,c,d | 2: b,c,d | 3: a,c,d | 4: a | 5:d

Question 1

Which of the following data sources are typically used to train large language models (LLMs)? Select three answers.

[ ] Images
[ ] News articles
[ ] Scientific papers
[ ] Textbooks

Question 2

A data scientist at a media company uses an LLM to better understand audience preferences. What potential issues should the data scientist check for when evaluating the LLM's output? Select three answers.

[ ] Predictive content
[ ] Insufficient output
[ ] Unfair bias
[ ] Inaccuracy

Question 3

When crafting a clear and specific prompt, which of the following elements should be included? Select three answers.

[ ] The task
[ ] Ambiguous details
[ ] A persona
[ ] The format

Question 4

An advertising executive is leveraging an AI tool to suggest slogans for a new drink. Which prompt provides the clearest instructions with the most context?

Create five potential slogans for a new drink. Use a casual tone. The drink has a tart taste and energizing herbs.
Slogans for new drink. Lots of energizing herbs. Really great taste.
List slogans for a new drink. We want to sell this drink to many customers. Come up with something interesting. Make sure that it's really catchy.
New drink slogans. Mention energizing herbs. Should be better than competitors' slogans.

Question 5

Fill in the blank: _____ is a technique that provides two or more examples in a prompt.

Sample-shot prompting

Low-shot prompting

Occasional prompting

Few-shot prompting

Module 4: Use Ai Responsibly

Responsible AI : Responsible AI is the principle of developing and using AI ethically with the intent of benefiting people and society, while avoiding harm.
- To ensure people are treated fairly and respectfully, AI users must be aware of the limitations of AI tools and commit to using them ethically. An AI user is someone who leverages AI to complete a personal or professional task, like editing copy for a marketing campaign, brainstorming ideas for a nonprofit fundraiser, or discovering more effective ways to use a particular technology.

Understand bias in AI

AI models are trained on data created by humans, so they consist of values and are subject to bias. They can also sometimes produce inaccurate results. Because an AI model is trained on a data set to recognize patterns and perform tasks, the model is only as good as the data it receives. The output from the AI tool may be affected by both systemic bias and data bias.
Systemic bias: Systemic bias is a tendency upheld by institutions that favors or disadvantages certain outcomes or groups.
- Systemic bias exists within societal systems like healthcare, law, education, politics, and more.
- Even if the people who design and train an AI model think they're using high-quality data, the data may already be biased because humans are influenced by systemic biases.
Data bias: Data bias is a circumstance in which systemic errors or prejudices lead to unfair or inaccurate information, resulting in biased outputs.
- Maybe you're developing a work presentation, you ask an AI image generator to create a photo of a CEO. All of the images generated appear to be white males. Based on this result, you might assume that all CEOs are white men. Obviously, this data is biased. Still, the more an AI model is trained with images of white men as CEOs, the more likely these models are to generate similarly biased outputs.
AI models are value-laden: Just as AI models reflect the biases of the data used to train them, they also reflect the values of the people who design them. In other words, AI models are value-laden.
- For example, perhaps an AI engineer wants to help create more sustainable ways to generate energy. The engineer could use AI to build a tool that allows energy suppliers to increase their use of renewable resources. In this case, the AI tool was created based on the idea that society can and should make the most of solar and wind power sources, which is a reflection of the engineer's values. This focus means that the AI tool is not intrinsically value-neutral. Other people may have different values about energy generation that are not reflected in this particular AI tool. Like most aspects of emerging technology, AI is not a perfect system. At present, it provides both opportunities and challenges, so using it responsibly requires critical thinking and an understanding about how data may be biased.

Identify AI harms

Allocative harm: An allocative harm is a wrongdoing that occurs when an AI system's use or behavior withholds opportunities or resources or information in domains that affect a person's wellbeing.
- For example, if AI tools don't provide the same information to everyone, some people may be denied access to education, healthcare, fair housing, or other opportunities. Maybe a property manager for an apartment complex uses an AI tool to screen applications for potential tenants. This AI tool uses the names and other identifying information on these applications to help conduct background checks. One applicant is deemed a risk because of a low credit score, so they are denied the apartment and they lose the application fee. Later, the property manager realizes the software had misidentified the applicant and ran a background check on the wrong person. In this example, the applicant has experienced an allocative harm because they were denied an opportunity and lost resources, both affecting their wellbeing.
Quality-of-service harm: Quality-of-service harm is a circumstance in which AI tools do not perform as well for certain groups of people based on their identity.
- When speech recognition technology was first developed, the training data didn't have many examples of speech patterns exhibited by people with disabilities so the devices often struggled to parse this type of speech, but this technology is still evolving.
Representational harm: an AI tool's reinforcement of the subordination of social groups based on their identities.
- For instance, the AI powering a language translation app might associate certain words with feminine or masculine traits, and choose gender specific translations based on those assumptions. Now, this is harmful because the result may be the erasure or alienation of social groups due to built-in biases.
Social system harm: This harm refers to macro-level societal effects that amplify existing class, power, or privilege disparities or cause physical harm as a result of the development or use of AI tools.
As AI-generated images become more realistic, there's concern about the spread of disinformation, including deep fakes. Deepfakes are AI generated fake photos or videos of real people saying or doing things that they did not do.
Deepfakes are AI generated fake photos or videos of real people saying or doing things that they did not do.
- An example of a social system harm might be if a deepfake of a school board candidate showed that person saying something they didn't say. If it went viral, causing them to lose the election, that would impact voters' perspectives, the way parents feel about their school district, and the community in general.
- Because there would be disinformation being spread on a large scale, this would be a social system harm.
- Fortunately, new technology is being created to detect Deepfakes. Some image generating tools are putting digital watermarks on AI-generated images and videos to indicate who created them.
Interpersonal harm: which is the use of technology to create a disadvantage to certain people that negatively affects their relationships with others or causes a loss of one's sense of self and agency.
- Sometimes people can share private information with an AI tool that could be misused by others, like locking someone out of an online account or surveilling them.

Security and privacy risks of AI

Privacy: Privacy is the right for a user to have control over how their personal information and data are collected, stored, and used.
Security: Security is the act of safeguarding personal information and private data, and ensuring that the system is secure by preventing unauthorized access.
Measures to protect privacy and security
- First, before you use an AI tool, be aware of its terms of use or service, privacy policy, and any associated risks.
- Next, don't input personal or confidential information.
  - To personalize your outputs, you can always edit the details later. Many AI tools use encryption and other measures to help protect your information, but you should always make sure to protect your privacy.
- Finally, stay up to date on the latest tools.
  - Knowing about new advancements in AI can help you understand risks as they come in. So if you plan on using AI frequently, make sure you're reading the latest articles from trusted news sources, scholarly and university publications, and subject matter experts.

Bias, drift, and knowledge cutoff

Social System Harm: Social system harm refers to macro-level societal effects that amplify existing disparities due to AI use.
Drift: Drift is the decline in an AI model's accuracy over time due to changes not reflected in the training data.
Bias: Bias in AI refers to systemic errors or prejudices in data that lead to unfair or inaccurate outputs.
Knowledge Cutoff: Knowledge cutoff is the point in time when an AI model was last trained, limiting its awareness of subsequent events or information.
Representational Harm: Representational harm involves AI reinforcing the subordination of social groups based on their identities.
Allocative Harm: Allocative harm occurs when AI systems withhold opportunities or resources, negatively impacting individuals' well-being.
Interpersonal Harm: Interpersonal harm is the disadvantage created by technology that negatively affects relationships and personal agency.

A thorough understanding of concepts in responsible AI—such as bias, drift, and knowledge cutoff—can help you use AI more ethically and with greater accountability. In this reading, you’ll learn how to use AI tools responsibly and understand the implications of unfair or inaccurate outputs.

Harms and biases

Engaging with AI responsibly requires knowledge of its inherent biases. Data biases are circumstances in which systemic errors or prejudices lead to unfair or inaccurate information, resulting in biased outputs. Using AI responsibly and being aware of AI’s potential biases can help you avoid these kinds of harms.

Biased output can cause many types of harm to people and society, including:

Allocative harm: Wrongdoing that occurs when an AI system’s use or behavior withholds opportunities, resources, or information in domains that affect a person’s well-being
- Example: If a property manager for an apartment complex were to use an AI tool that conducted background checks to screen applications for potential tenants, the AI tool might misidentify an applicant and deem them a risk because of a low credit score. They might be denied an apartment and lose the application fee.
- How to mitigate: Evaluate all AI-generated content before you incorporate it into your work or share it with anyone. Situations like the one in the example can be avoided by double-checking AI output against other sources.
Quality-of-service harm: A circumstance in which AI tools do not perform as well for certain groups of people based on their identity
- Example: When speech-recognition technology was first developed, the training data didn’t have many examples of speech patterns exhibited by people with disabilities, so the devices often struggled to parse this type of speech.
- How to mitigate: Specify diversity by adding inclusive language to your prompt. If a generative AI tool fails to consider certain groups or identities, like people with disabilities, address that problem when you iterate on the prompt.
Representational harm: An AI tool’s reinforcement of the subordination of social groups based on their identities
- Example: When translation technology was first developed, certain outputs would inaccurately skew masculine or feminine. For example, when generating a translation for words like “nurse” and “beautiful,” the translation would skew feminine. When words like “doctor” and “strong” were used as inputs, the translation would skew masculine.
- How to mitigate: Challenge assumptions. If a generative AI tool provides a biased response, like by skewing masculine or feminine in its output, identify and address the issue when you iterate on your prompt, and ask the tool to correct the bias.
Social system harm: Macro-level societal effects that amplify existing class, power, or privilege disparities, or cause physical harm, as a result of the development or use of AI tools
- Example: Unwanted deepfakes, ****which are ****AI-generated fake photos or videos of real people saying or doing things they did not say or do, can be an example of a social system harm.
- How to mitigate: Fact-check and cross-reference output. Some generative AI tools have features that provide sources for where information was found. You can also fact-check an output by using a search engine to confirm information, or asking an expert for help. Running a prompt through two or more resources helps you identify possible inaccurate output.
Interpersonal harm: The use of technology to create a disadvantage to certain people that negatively affects their relationships with others or causes a loss of their sense of self and agency
- Example: If someone were able to take control over an in-home device at their previous apartment to play an unwanted prank on their former roommate, these actions could result in a loss of sense of self and agency by the person affected by the prank.
- How to mitigate: Consider the effects of using AI, and always use your best judgment and critical thinking skills. Ask yourself whether or not AI is right for the task you’re working on. Like any technology, AI can be both beneficial and harmful, depending on how it’s used. Ultimately, it’s the user’s responsibility to make sure they avoid causing harm by using AI.

Drift versus knowledge cutoff

Another phenomenon that can cause unfair or inaccurate outputs is drift. Drift is the ****decline in an AI model's accuracy in predictions due to changes over time that aren’t reflected in the training data. This is commonly caused by knowledge cutoff, the concept that a model is trained at a specific point in time, so it doesn’t have any knowledge of events or information after that date.

For instance, a fashion designer might want to track trends in spending before creating a new collection. If they use a model that was last trained on fashion trends and consumer habits from 2015, the model may not produce useful outputs because those two factors likely changed over time. Consumer preferences in 2015 are very likely to be different from today’s trends. In other words, the model’s predictions have drifted from accurate at the time of training to less accurate in the present day due, in part, to the model's knowledge cutoff.

Several other factors can cause drift, making an AI model less reliable. Biases in new data can contribute to drift. Changes in the ways people behave and use technology, or even major events in the world can affect a model, making it less reliable. To keep an AI model working well, it's important to regularly monitor its performance and address its knowledge cutoffs using a human-in-the-loop approach.

To explore biases, data, drift, and knowledge cutoffs, check out the exercise

What Have Language Models Learned?

from Google PAIR Explorables. There you can interact with BERT, one of the first large language models (LLMs), and explore how correlations in the data might lead to problematic biases in outcomes. You can also check out other

PAIR AI Explorables

to learn more about responsible AI.

Checklist for using AI responsibly

Provision of High-Quality Information: The quality of AI outputs is influenced by the quality of the information provided; users should ensure their inputs are accurate and unbiased.
Clarity in Prompts: Being clear and specific in your prompts helps guide the AI tool away from generic outputs and toward the nuanced responses you desire.
Verification of Outputs: Users are responsible for reviewing and fact-checking AI-generated outputs to ensure accuracy and appropriateness.
Protection of Sensitive Information: Users must avoid including personal or confidential data in prompts, ensuring privacy and security when using public AI tools.

When you use a generative AI tool, you're in the driver's seat. The tool provides computational power, but you provide direction, judgment, and critical thinking. Using AI responsibly means being thoughtful and intentional with your prompts to ensure the quality, safety, and accuracy of the outputs.

This reading provides a set of key practices to help you make responsible choices each time you use AI.

Principles of responsible AI use

Applying the following four principles helps you use AI tools both responsibly and effectively.

1. Be clear and specific in your prompts

Vague or ambiguous prompts can lead to generic outputs. The more specific you are in your instructions, the more you guide the AI tool away from unhelpful assumptions and toward the nuanced output you actually want. Being specific is a key part of using AI responsibly.

2. Protect sensitive information

This is a crucial responsibility for every AI user. Public AI tools are not the place for personal, confidential, or sensitive data. Before entering any prompt, it’s a good idea to pause and ask yourself: Am I including data in my prompt that someone else might expect me to keep private?

Here are a few practical ways to protect sensitive information when prompting:

Use placeholders by referring to people, projects, or places with generic names.
Frame prompts around the task you need done, not the people involved.
Input only relevant context needed to complete the task, rather than entire documents.

Pro tip: Check out how to adjust your privacy settings and manage your conversation history in Gemini.

3. Provide high-quality information

The quality of an AI tool's output can reflect the quality of the references you provide. If you give the tool examples or data that are biased, outdated, or contain errors, the tool is likely to reproduce those same flaws in its response.

You should always try to provide inputs that are as accurate and unbiased as possible. Before using a piece of text or data as an example, ask yourself: Does this information represent a fair and balanced perspective? Is it from a source I trust?

By providing fair and factual inputs, you guide the tool toward creating a more responsible and accurate output.

4. Always verify the output

No matter how good your prompt is, the final responsibility for the output is yours. You are the essential human-in-the-loop who must be the final judge of accuracy, tone, and appropriateness.

Here is a simple workflow to apply this principle:

Review the output to ensure it meets your goals.
Fact-check any claims or data by cross-referencing the information.
Revise the content yourself by treating the AI-generated text as a first draft.

Pro tip: Some AI tools can help you fact-check their own responses. You can learn more about how to use the double-check feature in Gemini to evaluate its responses.

Ultimately, being a responsible AI user means combining the capabilities of the tool with your own critical thinking and judgment. By practicing each of these principles, you do more than just get better outputs—you contribute to a safer and more ethical use of AI.

Module 4 challenge

1: c | 2: C | 3: B | 4: A,C | 5: B,C,D

**1.**Question 1

An eco-friendly food distribution center is using AI to reduce waste. They ensure their AI models are used to benefit people and society while avoiding harm. What does this scenario describe?

Regenerative AI

Dynamic AI

Responsible AI

Progressive AI

**2.**Question 2

A homeowner receives an AI-powered robocall from a politician urging them to vote, but the call is fake and gives the wrong date, causing the homeowner to miss their chance to vote. What type of harm does this scenario describe?

Quality-of-service

Hallucination

Allocative

Knowledge cutoff

**3.**Question 3

A manufacturer starts making products faster thanks to a new AI tool. As a result, products arrive in stores 15% more quickly than they used to. After a few years, production slows because the AI model the tool uses is never trained on new data. What does this scenario describe?

Updates

Drift

Algorithms

Emergence

**4.**Question 4

Which two statements best describe data bias? Select two answers.

[ ] Data bias can exist within high-quality training data.
[ ] Data bias is a feature that enables AI models to fix errors in their programming.
[ ] Data bias is a circumstance in which systemic errors lead to inaccurate information.
[ ] Data bias ensures that AI models are trained on the latest information available.

**5.**Question 5

Which measures can you take to keep data private and safe when using AI tools? Select three answers.

[ ] Only use AI tools for work-related purposes.
[ ] Be aware of an AI tool's terms of use or service.
[ ] Stay up-to-date on the latest AI news.
[ ] Avoid inputting personal or confidential information.

Module 5: Stay Ahed of the AI Curve

Stay up to date with AI

Like many technologies, artificial intelligence (AI) is continually evolving, with new tools and breakthroughs announced almost daily. Staying informed about these changes is key to enhancing your work and optimizing your skillset. This guide provides a practical toolkit of strategies and resources to help you keep your AI knowledge up-to-date.

Create an AI Information toolkit

Forming a habit of staying informed doesn't have to be time-consuming. The goal is to create a personal, curated collection of sources that you can turn to regularly. Your toolkit can be as simple as a dedicated bookmark folder in your browser, a watchlist, or even a collection of links from trusted sources in a note-taking app. Use any combination of the following strategies to start building your own toolkit:

Curate a reading list: Find a few trusted newsletters or blogs that summarize the week's biggest AI news. A good strategy is to pair an aggregator newsletter like The Neuron with primary sources like the official Google DeepMind Blog for research breakthroughs and The Keyword for the latest Gemini product news.
Follow the conversation: A lot of AI news and discussion happens in real-time on social and professional networks. Follow the official Google page on LinkedIn, or Stanford's Institute for Human-Centered Artificial Intelligence HAI, where industry leaders, and prominent researchers share the latest developments as they happen.
Listen and watch: Use your commute or downtime to learn by listening to podcasts or watching videos. A great place to start is the Hard Fork podcast from The New York Times, which offers insightful weekly conversations about the latest in AI. Resources like these are great for understanding the context and stories behind the headlines.
Experiment with new tools: Perhaps the most effective way to understand advancements in AI is to use them. When a new AI tool or feature is released, take a few minutes to try it out. This hands-on experience will give you a much deeper and more practical understanding of a tool's capabilities than just reading about it.

The world of AI can feel like it's moving at lightning speed. Staying informed isn't about knowing everything—it's about creating a simple, consistent learning habit. Curating your own toolkit of trusted resources helps turn the latest AI developments into a manageable stream of inspiration. This approach empowers you to discover new possibilities and stay confident in a rapidly changing world.

Activity: Evaluate a new AI tool

Scenarios

First, read through the following options. Then complete the activity using the option that best suits you.

Option 1: Evaluate a new AI tool for a specific task

Think of a common task you perform for work, school, or even a personal project that involves creating content. For example, it could be drafting a professional email, brainstorming ideas for a presentation, writing a section of a report, or generating a logo for a website. You want to search for a relevant AI tool and evaluate it to determine whether it can help you complete that task.

Option 2: Evaluate Google AI Studio as a marketing tool

Imagine you’re a marketing manager for a company that builds audio equipment, like speakers and headphones. You recently read an article about Google AI Studio, and you think it could help you create digital ads more efficiently, like promotional videos, audio advertisements, and more. You decide to evaluate the tool to decide if it's a good fit.

Once you’ve chosen an option, follow the steps in this activity to practice exploring and assessing new AI tools.

Step 1: Define your goal

For option 1

Take a moment to picture your ideal outcome of using an AI tool for the task you've chosen. Consider, what does a successful result look like to you?

For example:

Drafting a report? Success could be an easy to scan summary with three key takeaways.
Generating a logo? You might want a professional design in a high-resolution format that's ready for the web.
Creating a voice-over? A great outcome might be a clear audio recording that brings your script to life.

Make a mental note of your goal, and when you're ready, proceed to Step 2 to find an AI tool to help you achieve it.

For option 2

Imagine that, in your role as a marketing manager for an audio equipment company, you're managing a big ad campaign that's coming up, and you need marketing materials for podcasts.

Your goal is to use an AI tool to see if you can efficiently create:

A compelling audio ad: A great result is a clear, high-quality audio recording of a commercial script that's ready for a podcast.

You'll test out Google AI Studio to determine whether it can help you create the ad content you need.

Step 2: Explore new tools

For option 1

Now that you’ve defined the goal for your task, it's time to explore AI tools that can help!

Identify at least one tool that you want to evaluate. To help you decide, do some quick research online to familiarize yourself with the tool. For example, you might:

Visit the tool's website to learn about its features
Read an article online that summarizes the tool's capabilities
Watch a YouTube video that highlights the tool's capabilities.

This will help you understand what the tool can do before you invest the time to test it out for yourself.

Note: If you have trouble finding a tool, refer to generative AI tools for workplace tasks for a list of tools that might be useful.

For option 2

Now it's time to see if Google AI Studio is the right tool for the job!

Before diving in, do some quick research to get a feel for what the tool can do. You might:

Review AI Studio's quickstart guide to learn the basics
Explore the developer documentation to get an overview of its key features
Watch a quick tutorial video that shows it in action

This initial research will give you a solid understanding of the tool's capabilities before you invest time in creating your ads.

Step 3: Evaluate the tool

For option 1

It's time to go hands-on with the tool you identified in Step 2. Here, your goal is to get a feel for how the tool performs on the task you defined earlier.

As you experiment, evaluate the tool based on these key questions:

Does the tool work as you expected? Is the output high-quality?
Does the tool align with any relevant security and third-party tool policies?
Does it have the specific features you need to achieve your goal?

Pro Tip: If you're evaluating a few different tools, keep a simple scorecard for each one. This will make your final decision much easier.

For option 2

It's time to put Google AI Studio to the test on your marketing challenge. Run a series of tests to get a feel for the tool and how it works.

Audio ad: Open AI Studio's s audio generator and do the same. Write a clear prompt to generate the voice-over for your podcast ad. Listen to the result, then iterate on the prompt to get a clean, high-quality recording.

After experimenting with the tool for a short while, make a few notes on how you thought it performed.

Step 4: Assess the tool’s suitability

It's time to bring it all together. You’ve set a clear goal and explored a new AI tool. Now, let's decide if it's the right fit for you.

Compare your notes from Step 3 against the goals you outlined in Step 1. Ask yourself:

Did the tool help you accomplish your task effectively?
Can you see yourself incorporating this tool into your work?
What’s your recommendation—is this tool a "Yes," a "No," or a "Maybe with more testing"?

At this stage, remember, a "No" can be just as valuable as a "Yes." The goal of this process is to evaluate tools for your specific needs. Every test is an opportunity to learn that gives you more clarity on what you're looking for!

Take inspiration from AI innovation

To start innovating with AI, consider what tasks you need to accomplish. For example, think about what slows down your work. Then think about how incorporating AI into your workflow may help you meet your needs
multimodal model: A multimodal model is an AI model that can accept and learn from multiple types of input, such as images, video, or audio.
- AI tools that incorporate multimodal models can perform additional tasks beyond what a model that analyzes just one type of data can do.

Activity: Plan for future opportunities with AI

Before you get started on the activity, review the prompt framework: Task (including Persona and Format), Context, References, Evaluate, and Iterate. Be sure to apply these prompting best practices when you craft your prompts.For more guidance, review the reading on prompting best practices.
Remember that providing human oversight is crucial when using an AI tool, so always be sure to review AI-generated output and give additional instructions to the tool as needed.

Scenario: Create your AI action plan

Take a moment to reflect on how you might apply AI tools to your daily life. Think about tasks where generative AI tools could save you time, enhance your creativity, or help you solve problems more efficiently. Consider which tools from this course seemed most relevant to your needs. As you work through the following steps, prompt the AI tool of your choice to develop your action plan, focusing on realistic opportunities where AI can add value.

Important: Throughout this activity, make sure you enter each prompt in the same chat in the AI tool because the prompts build on each other.

Step 1: Determine next steps

Prompt the AI tool to help you brainstorm ways to use it in your daily life.

You might phrase your prompt like one of these examples:

At work, I often need help _________. How can an AI tool help me with that?
Every day, I have to _________. How can I leverage an AI tool to help me _________?
A common challenge I encounter is _________. How can I use an AI tool to overcome that challenge?

Review the output that the AI tool creates. Does it come up with actionable ideas? If not, enter another prompt explaining why, then ask it to revise the output. Repeat this process until you’re satisfied with the final output.

Step 2: Set goals for yourself

In the same chat, enter this prompt into the AI tool:

What are some specific goals I can set to continue developing my AI skills? Are there actions I can take to accomplish those goals?

Review the output that the AI tool creates. Does it identify goals that are practical for you to accomplish? If not, enter another prompt explaining why, then ask the tool to try a different approach. Repeat this process until you’re satisfied with the output.

Step 3: Create your AI action plan

In the same chat, enter this prompt into the AI tool:

Create an AI action plan just for me. Format the plan in a checklist. It should include sections with the next steps I want to take and the specific goals I want to accomplish. The plan should also include a timeline.

Review the action plan that the AI tool creates. Does it include all the information you want it to? If not, enter a follow-up prompt explaining what’s missing, then ask it to revise the output. Repeat this process until you’re satisfied with the final output. Then, download the plan or copy and paste it into a separate document for future reference as you continue to use the AI tool.

Take the next step with Google Prompting Essentials

Ready to put AI to work for you? In this lesson, you’ll get an exclusive preview of Google Prompting Essentials, a self-paced program developed by AI experts at Google that teaches you how to get the most out of generative AI tools in 5 easy steps. You’ll apply everything you’ve learned in Google AI Essentials—and more—to unlock the full potential of AI.

Through hands-on activities and real-world examples, Google Prompting Essentials can teach you how to get the most out of AI to:

Save time on daily tasks: Craft tailored emails, brainstorm creative content with ease, build tables and trackers effortlessly, and quickly summarize documents.
Uncover and share powerful insights: Identify patterns in data, create compelling visuals, and even rehearse presentations.
Tackle complex projects: Transform abstract ideas into actionable steps, using AI to role-play conversations, and get expert feedback.

In 5 easy steps, you’ll learn how to craft effective prompts. You’ll also learn more about responsible AI practices and start building a library of reusable prompts so you’re not prompting from scratch every time that you need a little help from AI. And when you complete the courses, you’ll earn a certificate from Google to share with your network and employer, just like with Google AI Essentials.

Explore Google Prompting Essentials, where you can put your expertise into practice!

Use the 5-step prompt framework

A good prompt follows a simple framework, task, context, references, evaluate, and iterate.
Task: You need to describe the task you want the generative AI tool to help you with. Now, this should include a persona and a format preference, so that the task is specific.
- Persona refers to what expertise you want the gen AI tool to draw from.
  - You can ask the tool to take on a persona like a professional speech writer or a marketing executive with 15 years of experience. Or you can ask it to create output for a specific audience, a customer, or even your manager. You can be as detailed as you'd like when adding a persona to your task.
- Format refers to how you want the output to appear, whether that's a bulleted list, short sentences, or a table.
Context or the necessary details to help the gen AI tool understand what you need from it.
- This is the difference between writing.
- Give me some ideas for a birthday present under $30 and give me five ideas for a birthday present. My budget is $30. The gift is for a 29-year-old who loves winter sports and has recently switched from snowboarding to skiing.
References for the gen AI tool to use while creating its output.
Evaluate: Once you have your output, it's time to evaluate. Ask yourself if the input you provided gave you the output you needed.
Iterate: If you evaluate your output and determine that you're not getting what you need, you can try again by adding more information or tweaking your prompt.
The order of how you construct a prompt is less important than this substance of the prompt itself. As long as you're thoughtfully creating, really, excellent inputs (TCREI) , your outputs should be great.

Module 5 Challenge

1:D | 2: C| 3: A | 4: A,C,D | 5: D

**1.**Question 1

An application developer is considering investing in a generative AI tool that can help them debug code. How might they evaluate the AI tool for this task?

Assess the tool's ability to create engaging web content.

Investigate the tool's marketing materials to determine its popularity.

Verify that the tool can generate speech from text.

Ensure that the AI tool can generate code snippets that are accurate.

**2.**Question 2

Fill in the blank: A(n) _____ model is an AI model that can learn from various modalities of input, such as video and written text.

speech-recognition image-classification multimodal uniform

**3.**Question 3

Fill in the blank: _____ tools can perform tasks such as summarizing information and creating images from text.

Generative AI Versatile AI Responsible AI Coding AI

**4.**Question 4

An IT manager wants to ensure that they remain knowledgeable about new advancements in AI. What strategies should the manager employ to stay up-to-date? Select three answers.

[ ] Network with IT professionals who are interested in AI.
[ ] Dedicate all their resources towards internally developing AI-powered solutions.
[ ] Read articles about how other professionals are applying AI to address IT challenges.
[ ] Engage with new AI tools as they become available.

**5.**Question 5

Which of the following strategies is effective for leveraging AI in your work?

Use AI to secure confidential projects Implement AI to automate the way you address sensitive issues Replace outdated software with AI-powered software Identify and address challenges that AI is capable of handling

Google AI Essentials - Lecture from 'Coursera'

Google AI Essentials

Certification: https://coursera.org/share/3caa5803f5adcfc2fdadfab2f4ffede6

Module 1: Introduction to AI

Overview of AI tools

A Guide to AI and ML

Generative AI and machine learning

Approaches to machine learning

Rule-based AI: A different approach

Resources for more information

Understand the capabilities and limitations of AI

Activity: Use AI to create an email

Scenario: Draft an email

Step 1: Prompt the AI tool to draft an email

Step 2: Review the output

Step 3: Revise the Email

Use AI as a collaborative tool

Module 2: Maximize productivity with AI tools

Types of generative AI Tools

Understand how AI tools work

AI models and the training process

AI tools and AI models

The process of training AI models

Transform your work with generative AI

Generative AI tools for workplace tasks

Text and content generative AI tools

Code-generative AI tools

Image- and media-generative AI tools

Integrating AI tools into your workflow

Activity: Use generative AI to develop content and images for social media

AI tool access

Option 1: Develop a social media post of your own

Option 2: Develop a social media post for a pet care company

Step 1: Generate text for your social media post

Step 2 : Generate an image for your social media post

Leverage the human-in-the-loop approach to AI

Determine if generative AI is right for the task

Use Gemini in Google Docs, Slides, Sheets, Meet, and Gmail

Introduction to Gemini in Workspace apps

Google Docs

Gmail

Google Slides

Google Sheets

Google Meet

Activity: Increase productivity with the help of Gemini for Google Workspace

AI tool access

Note: Before you use Gemini in Workspace apps, review the following information:

Option 1: Explore Gemini in Workspace apps in your daily life

Option 2: Manage a bookstore with help from Gemini in Workspace apps

Generate insights from multiple sources with NotebookLM

An AI tool with source grounding

Explore NotebookLM with Shakespeare

Wrap-up

Resources for more information

Module 2 Challenge (Quiz)

Module 3: Discover the Art of Prompting

Prompting best practices

Provide necessary context

Include references as examples

Evaluate your output

Iterate for better results

Activity: Write an effective prompt for Gemini

Step 1: Write a prompt

Step 2: Evaluate your prompt

Step 3: Keep iterating

Prompts for different purposes

Content creation

Summarization

Classification

Extraction

Translation

Editing

Problem-solving

Activity: Iterate on prompts to find solutions

Scenarios

Option 1: Solve a workplace problem of your own

Option 2: Solve a scheduling dilemma problem

Step 1: Prompt to generate solutions

Step 2: Evaluate the output

Step 3: Iterate for better results