Innovating with Google Cloud Artificial Intelligence · Topic 1 of 4
AI, Machine Learning, Generative AI and Agents
Cloud Digital Leader definitions of AI, ML, generative AI, data analytics and BI, how agentic AI is reshaping work, and the business problems ML can solve.
- 4 min
- AI is the field; ML learns from data; gen AI creates content; agents plan and act.
- Analytics and BI explain what happened; ML predicts; gen AI generates.
- ML suits problems where rules are hard to write or data is large, and decisions must scale.
What it is
Artificial intelligence (AI) is the broad field of making machines perform tasks that normally need human intelligence. Machine learning (ML) is the part of AI where systems learn patterns from data instead of following hand-written rules. Generative AI is ML that creates new content. Agentic AI uses AI agents that plan and take actions toward a goal. The exam expects you to tell these apart from data analytics and business intelligence, and to know which business problems ML solves well.
How it works
The definitions
| Term | What it means | Example |
|---|---|---|
| Artificial intelligence (AI) | Machines performing tasks that usually need human intelligence, such as understanding language or recognizing images | A system that reads invoices |
| Machine learning (ML) | A way to build AI by training models on data so they learn patterns and make predictions | Predicting which customers will cancel |
| Generative AI (gen AI) | Models that generate new content: text, images, code, audio, video | Drafting a product description |
| Data analytics | Examining data to find patterns and answer questions, often about what happened and why | Analyzing why sales fell last quarter |
| Business intelligence (BI) | Tools and practices that present data in reports and dashboards for decisions | A weekly sales dashboard |
Analytics and BI mostly describe the past and present. ML predicts and automates. Gen AI creates.
If the system creates something new (an email, an image, a summary), it is generative AI. If it predicts a value or category, it is classic ML.
Agentic AI
An AI agent is a system that pursues a goal: it can reason about the steps, use tools and data (search a catalog, call an API, update a record) and act, checking its results along the way. Agentic AI is reshaping how people work:
| Area | How agents change it |
|---|---|
| Workforce productivity | Agents find information, draft documents and complete routine multi-step tasks for employees |
| Customer support | Agents resolve requests end to end, such as changing a booking, not just answering questions |
| Sales experiences | Agents qualify leads, prepare account research and personalize offers |
| Product innovation | Agents help teams prototype, write code and test ideas faster |
| Operations | Agents monitor systems and processes and take corrective actions |
| Research | Agents gather and summarize large amounts of information |
The key difference from a chatbot: a chatbot answers, an agent acts toward an outcome.
Business problems ML solves
ML is valuable when rules are hard to write or the data is too large for people to analyze.
| Problem type | Business value | Example |
|---|---|---|
| Replacing or simplifying rule-based systems | Thousands of hand-written rules become one model that adapts to new patterns | Spam and fraud detection |
| Deriving insights from large datasets | Find patterns in structured and unstructured data people would miss | Analyze millions of support tickets for common issues |
| Scaling business decisions | Make consistent decisions for every customer, instantly | Personalized recommendations for millions of shoppers |
Common ML use cases: forecasting (demand, sales), classification (spam or not, fraud or not), recommendation, anomaly detection (unusual transactions or sensor readings), image and document understanding, translation and speech recognition.
A worked scenario
An airline uses BI dashboards to see last month's delays, analytics to find which routes cause them, an ML model to predict tomorrow's delays and rebook passengers in advance, generative AI to draft apology messages, and an AI agent that rebooks a passenger's flight, issues a meal voucher and updates the booking system without a human handling each step.
How the exam asks about this
- A description is given and you choose the term (ML, gen AI, BI, analytics).
- "Which capability completes a multi-step task using tools?" Agentic AI.
- "A rule-based fraud system cannot keep up with new fraud patterns." Replace it with ML.
- "Which industry change does agentic AI bring?" Productivity, support, sales, product, operations, research.
What to remember for the exam
- AI is the field; ML learns from data; gen AI creates content; agents plan and act.
- Analytics and BI explain what happened; ML predicts; gen AI generates.
- ML suits problems where rules are hard to write or data is large, and decisions must scale.
- Agentic AI changes productivity, customer support, sales, product innovation, operations and research.
Do not choose ML for a problem a simple, stable rule solves well. ML adds value when patterns are complex or changing.
Common mistakes
Treating AI and ML as identical. ML is one way to achieve AI.
Calling a dashboard AI. Dashboards are business intelligence.
Thinking agents are just chatbots. Agents take actions with tools to reach a goal.
Check yourself
What is the difference between machine learning and generative AI?
Machine learning learns patterns to predict or classify; generative AI is ML that creates new content such as text or images.
What makes an AI system "agentic"?
It plans and takes actions toward a goal, using tools and data, rather than only answering.
Name a business problem that ML solves better than rules.
Detecting fraud or spam whose patterns change too fast for hand-written rules.
Topic 1 of 4 in Innovating with Google Cloud Artificial Intelligence
Test yourself on Innovating with Google Cloud Artificial Intelligence
The full Google Cloud Digital Leader question bank has 512 questions, with timed practice exams and an explanation for every answer.
This study guide is independent and is not affiliated with, endorsed by, or sponsored by Google Cloud. Vendor names, certification names and exam codes are trademarks of their respective owners and are used for identification only.