Discover how AI works
This course content explains the basics of AI, focusing on how AI tools work and the role of machine learning in powering them.
AI Tools and Their Applications
- AI tools are software powered by AI that automate or assist with various tasks, such as video recommendations on streaming platforms, GPS route suggestions, and real-time translation.
- Companies use AI tools to improve operations and product quality, enhancing user experiences across many industries.
Machine Learning as the Foundation of AI Tools
- Machine learning (ML) is a subset of AI that enables programs to analyze data and make decisions or predictions.
- ML programs are trained using training sets—collections of data examples that teach the AI how to recognize patterns and respond appropriately.
Training Data and Potential Bias
- The effectiveness of ML depends heavily on the quality and relevance of the training data provided.
- Bias in training data can lead to inaccurate or unintended outputs, such as an AI tool misidentifying ripe apples if it was only trained on certain types of apples.
Overall, machine learning is a powerful technique that drives many AI tools, but careful attention to training data is essential to ensure accurate and fair AI performance.
Machine Learning: A Simple Explanation
Imagine teaching a child to recognize ripe apples. You show them many pictures of apples—some ripe, some not—and explain the differences. Over time, the child learns to spot ripe apples even in new pictures they’ve never seen before. Machine learning works in a similar way for computers. Instead of a child, we have a computer program, and instead of pictures, we give it lots of data called a training set. This training set helps the program learn patterns and make decisions, like identifying ripe apples in a factory.
Just like the child might get confused if they only see one type of apple, machine learning programs can also make mistakes if their training data is limited or biased. For example, if the program only sees red apples, it might not recognize green ones as ripe. So, the quality and variety of the training data are very important to help the AI tool work well and make accurate decisions.
Real-life examples
Certainly! Here’s a real-life example related to the key concept of AI tools powered by machine learning (ML) and training data from the course material:
Example: AI in Email Spam Filtering
Relevance: This example shows how AI tools use machine learning to improve everyday digital experiences, which is relevant to understanding how AI impacts common tasks.
- Explanation:
- Email providers use AI tools to automatically detect and filter out spam emails.
- To build this AI tool, designers train an ML program with a training set containing thousands of examples of spam and non-spam emails.
- The ML program learns to identify patterns and features typical of spam emails, such as certain keywords, sender addresses, or formatting.
- Once trained, the AI tool can accurately classify new incoming emails as spam or legitimate, even if those emails were not part of the original training set.
- This helps users by keeping their inboxes clean and reducing the risk of phishing or malicious content.
- However, if the training data is biased or incomplete (e.g., missing examples of certain types of spam), the AI tool might misclassify emails, either letting spam through or wrongly marking important emails as spam.
Foundations of generative AI
introduces generative AI, a type of AI that creates new content such as text, images, or media using natural language input.
Generative AI Basics
- Generative AI tools accept inputs like text, speech, images, or videos and generate outputs in various media forms.
- These tools allow users to interact with AI using natural language, making content creation accessible without specialized skills.
Applications of Generative AI
- Generative AI can assist in creating marketing materials, drafting emails, brainstorming ideas, and improving decision-making.
- It complements human skills by enabling iterative refinement through additional instructions until desired results are achieved.
Conversational AI and Practical Use
- Conversational AI tools, like Google’s Gemini, process text requests and generate helpful responses for creativity and productivity.
- Gemini can brainstorm ideas, provide detailed explanations, and support tasks across various professional fields, enhancing work efficiency.
Generative AI: Creating New Content with AI
Generative AI is a type of artificial intelligence that can create new things like text, pictures, or sounds. Imagine you have a magic helper who listens to what you say or write and then makes something new for you based on that. For example, if you want a poster for your new business but don’t have a designer, you can tell the AI what you want, and it will create a poster for you. If you don’t like it, you can give more instructions, and it will try again until it gets it right.
Think of generative AI like a creative partner who understands your natural way of talking or writing and helps you come up with ideas, write emails, or even plan events. It’s like having a helpful friend who can brainstorm with you and make your work easier and more fun. For example, a conversational AI tool can suggest fun activities for a team retreat or help you write a quick reply to an email. This technology is useful in many fields, from healthcare to education, making tasks faster and more creative.
Real-life examples
Certainly! Here’s a real-life example related to the concept of generative AI and its use with natural language, as explained in the course content:
Example: Marketing a Small Business with Generative AI
- Imagine you own a small bakery and want to promote a new line of cupcakes. You don’t have a dedicated marketing team or graphic designer.
- Using a generative AI tool, you type a simple instruction like “Create a colorful promotional poster for my new cupcake flavors, including chocolate, vanilla, and strawberry.”
- The AI processes your input and generates a visually appealing poster with images and text highlighting your cupcakes.
- If you want changes, you can provide more instructions, such as “Make the background pastel pink and add a discount offer.”
- This helps you quickly create marketing materials without needing specialized skills, saving time and boosting your business visibility.
Why this example is relevant:
- It directly connects to the course content about generative AI generating new content from natural language input.
- It shows how generative AI complements human skills by handling creative tasks efficiently.
- It illustrates practical use in a common industry scenario, making the concept easier to understand and apply.
A guide to AI and ML
Generative AI tools can create new and original content, from text and images to music and code. But how do these tools produce such creative and relevant results? The answer lies in the relationship between artificial intelligence (AI) and machine learning (ML).
Understanding how AI and ML work together is a valuable professional skill. It can help you navigate new technologies as they transform the workplace, enabling you to contribute to AI-driven projects or even lead your own AI initiatives.
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 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
Complete activities with Google AI Pro at no cost
In this program, you’ll do more than just learn about AI. You’ll actually build with it. You’ll be taught by a team of Google AI experts and practitioners who use these tools everyday. They’ll serve as your guides as you explore different ways to apply AI to your work.
But before you can build like an AI professional, you’ll need professional-grade tools.
Get 3 months of Google AI Pro at no cost
To help you get the most out of the program, we’re providing a complimentary three month trial subscription to Google AI Pro. While optional and not required to complete your activities, or earn your certificate, it gives you higher access to a suite of our most capable AI models and features.
How to access the extended trial:
Proceed to the next item
Select “I agree to use this app responsibly”
Click on the “Launch App” button
A new browser window will open and a unique link will populate
Click on your unique link (do not share this link with others, it is assigned to you only)
Click on start trial
Make sure you are signed into Gemini with your Google account to complete your activities with Google AI Pro
Tips and reminders:
Eligibility requirements: The trial is only available in countries where Google AI Pro is offered, can only be redeemed once, and has an age requirement.
Google account requirement: The promotional code is designed for personal Google Accounts. If you are currently signed into a Google Workspace account, please switch to your personal account to redeem the offer.
Using AI at work: Always remember to follow your organization’s AI guidance when using these tools for work related tasks.
Using AI at school: Before using Google AI Pro or other generative AI tools to complete your labs, check your institution’s AI policy and ensure your use of AI aligns with your school’s data privacy and compliance guidelines (e.g., FERPA). Do not input institutional data or private records into AI tools unless they are part of your university’s approved data ecosystem.
Ready to get started? Proceed to the next item.
Note: While a Google AI Pro subscription is not strictly required to complete the program, it is highly recommended to enhance your learning experience and unlock the full potential of the hands-on activities. See all terms and conditions.
