[NEW] Be a Generative AI Leader – Google Cloud Certification

Define artificial intelligence and distinguish strong general AI from narrow weak AI, with examples like self-driving cars, spam filters, and recommendation systems.

Explore the differences between AI, ML, and generative AI with Gemini, a Google chatbot. Craft prompts, generate text and images, and discuss access, limitations, and foundational models.

Learn how machine learning differs from traditional programming by training algorithms with millions of data examples to build models that predict outcomes, like house prices or handwritten digits.

Learn the difference between features and labels in machine learning, and how models use inputs to make predictions, illustrated with house price, car price, spam, and loan examples.

Learn the five stages of the machine learning lifecycle—data ingestion, preparation, training, deployment, and management—and how to collect, clean, train, deploy, and monitor models.

Explore structured data versus unstructured data, with examples like employee tables, flight bookings, and JSON or Excel formats to show how organization aids searchability.

Differentiate labeled data from unlabeled data, showing how tagged inputs enable supervised learning while raw data supports unsupervised or semi-supervised learning, for pattern discovery and clustering.

Distinguish supervised learning from unsupervised learning by using labeled data to train models for predicting outcomes, and by uncovering hidden patterns in unlabeled data through clustering.

Explore reinforcement learning as learning by trial and error through actions, feedback, and rewards and penalties, illustrated by game AI, video suggestions changing in real time, and robot arms.

Compare supervised, unsupervised, and reinforcement learning through real-world examples, including predicting house prices with labels, clustering customers, and learning from feedback via trial and error.

Navigate the AI turmoil with a pragmatic approach, learning to use AI through APIs, consume and integrate solutions, and boost productivity without needing deep AI engineering.

Learn how generative AI differs from AI and machine learning and how it creates new content, from text generation and emails to code, images, and media.

Learn how generative AI uses foundation models pre-trained on massive data. Recognize their adaptability to a wide range of tasks, and the need for huge infrastructure and GPUs and TPUs.

Explore foundation models, specifically large language models trained on vast diverse text to predict the next token, with temperature tuning for creativity, and applications like chat, summarization, translation, and Q&A.

Explore diffusion models, often built into foundation models, that start with random noise and iteratively refine prompts to produce images, audio, or video.

Explore foundation models through scenarios, distinguishing large language models and diffusion models. Learn how language models handle text and diffusion models generate images, audio, or video from noise.

Discover how Gemini, a multimodal foundation model in the Google ecosystem, understands and generates text, images, code, audio, and video.

Explore Imagen, a Google foundation model trained on images and their text descriptions, enabling text-to-image generation, image editing from prompts, and image content understanding for social graphics and illustrations.

Chirp, a foundation model in the Google ecosystem, trains on multilingual audio data to enable speech recognition, voice-based interaction, real-time transcription, and audio translation across languages.

Discover Google's Veo foundation model for video generation, enabling text-to-video and image-to-video workflows with audio from prompts.

Explore Google AI Studio to prototype with foundation models in a code-free interface, log in to access the model playground and generate apps and code for Gemini APIs.

Explore Google AI Studio to interact with foundation models like Gemini, run prompts, and generate Python code with a Gemini API key to execute text, image, and video prompts.

Explore Google's Gemma foundation model family, a lightweight, open, state-of-the-art option for text, image, and code tasks, including Gemma Code and Paligemma for visual data processing.

Explore the key limitations of foundation models, including data dependency, knowledge cutoff, hallucinations, bias and fairness, and edge cases in real-world applications.

Learn to reduce hallucinations and improve accuracy by grounding AI models to trusted data, such as product documentation, using retrieval augmented generation to pull internal knowledge base information into prompts.

Compare zero shot, one shot, and few shot prompting by including examples to guide model responses and stabilize consistent JSON outputs.

Assign a persona to the model to set tone and style through role prompting. Configure system instructions in Google AI Studio to craft empathetic, customer service style responses.

Learn how prompt chaining breaks complex tasks into smaller linked prompts, guiding a model step by step to improve accuracy in multi-step reasoning and task execution.

Explore chain of thought prompting to guide models through step-by-step reasoning, improving accuracy and explainability with prompts and real-world examples like math, speed, and debugging.

Explore ReAct prompting, a technique that alternates reasoning and acting to perform actions like searches, reflect on intermediate steps, and produce grounded, tool-augmented answers.

Explore practical prompting techniques, including role prompting and chain-of-thought reasoning, across customer service, math word problems, career counseling, and medical diagnosis scenarios.

Explore how to tune generative ai models by adjusting output token limits, top-p, and temperature to control randomness and creativity in responses.

Explore how adjusting temperature and Top-p affects model outputs across scenarios, from precise math answers to creative poetry, formal emails, and fantasy name brainstorming, including token limit considerations.

Explore edge computing with Gemini Nano, a compact AI model designed for on-device use that delivers fast, private, offline AI and real-time decisions.

Gemini integrates with Google Cloud and Google Workspace to enable natural language interactions, letting you create resources, generate SQL, and explore data in BigQuery, Looker, and other apps.

Explore how Gemini's up to 1 million token context window enhances memory across long conversations, books, codebases, and videos, enabling processing and reasoning on large documents.

Configure Gemini safety settings to guard against harmful or biased responses and prevent misuse by adjusting levels for harassment, hate, and dangerous content in aistudio.google.com advanced settings.

Learn to build custom experts with gems in Gemini, using pre-made options like career guide, brainstormer, and coding partner, and configure context, prompts, tone, and grounding with your own files.

Explore Gemini gems to access pre-made options like Brainstormer gem, career guide gem, and coding partner gem, and learn to create, configure, publish your own gem with prompts and files.

Explore grounding and retrieval augmented generation, or RAG, connecting ai output to verifiable sources with real-time external knowledge, citations, and reduced hallucinations.

Explore NotebookLM, built on Gemini models and grounded in your uploaded documents, delivering reliable outputs with citations; generate one-click summaries, FAQs, briefing docs, and audio or podcast-ready content.

Explore NotebookLM by uploading PDFs, text, Markdown, audio, or Drive links to base responses on your sources and generate study guides, mind maps, FAQs, and audio overviews.

Explore practical NotebookLM use cases, from uploading employee handbooks to generate summarized FAQs, briefing executives from market reports, to creating study guides, slide deck summaries, and audio earnings overviews.

Explore how Google's ai first strategy and innovations from search and translate to voice search, TensorFlow, tpus, and transformer models like Gemini shape modern ai and responsible ai practices.

Explore Google Cloud's 200-plus compute, storage, networking, database, and AI services, with Vertex AI to build, deploy, and fine-tune generative AI models, backed by encryption and IAM.

Google Cloud enables business-focused AI deployments by handling infrastructure, offering access to cutting-edge models, open standards like TensorFlow and Kubernetes, and scalable, secure tools.

Explore why Google Cloud enables multi-cloud flexibility, scalable infrastructure, and access to foundation models like Gemini Imogen, with responsible AI guidance and explainable AI tools.

Explore the traditional Google Cloud ML landscape before generative AI, using APIs like Natural Language API, Vision API, Speech API, Video Intelligence API, plus AutoML and Vertex AI training.

Explore Vertex AI, Google Cloud's AI platform, to build, train, deploy, and scale traditional ML and generative AI with end-to-end MLOps, model registry, Model Garden, and Studio tools.

Explore Vertex AI Studio to test, tune, and deploy enterprise-grade generative AI with prompt management, grounding options, and model tuning, plus a prompt gallery and model garden.

Compare Google AI Studio and Vertex AI Studio to see how beginners can quickly explore foundation models, while enterprises deploy, tune, and govern production-grade solutions.

Explore AI agents that observe, reason, decide, and act as goal driven assistants. See how they automate tasks across customer support, banking, HR, data analytics, and code development.

Explore conversational agents that understand natural language, identify intents, fetch data, and craft responses with tools. Examine workflow agents that automate multi-step processes, plan and execute actions, and deliver results.

Explore how AI agents work by examining four core parts—persona, memory, model, and tools—along with extensions, functions, data stores, and plugins that enable external actions.

Build agents in cloud using Vertex AI Agent Builder, writing prompts or code and grounding them in enterprise data; deploy with Vertex AI Agent Runtime and explore Agent Garden samples.

Google Agent Space centralizes enterprise knowledge, enabling custom AI agents to access, understand, and act on data across Google Drive, JIRA, Confluence, and SharePoint, and integrate into internal dashboards.

Unlock enterprise search with Vertex AI Search, a fully managed Google Cloud solution that combines websites, PDFs, documents, and BigQuery data to ground generative AI with RAG.

Explore Google's customer engagement suite, including conversational agents, agent assist, conversational insights, and contact center as a service. Learn how these tools enable multi-channel, AI-powered support and live agent collaboration.

Explore Google's customer engagement suite scenarios, featuring 24/7 chat and voice support, agent assist with real-time suggestions, and conversational insights. Consider cloud-based contact centers that unify chat, voice, and email.