Revisions
AWS exam guides are periodically reviewed and updated to ensure that our certification exams test skills and AWS services and features that are relevant for the job role(s) that a certification is designed to target. Exam guide updates will be published approximately one month before updates will be reflected on your exam.
Change History
| Version | Publication date |
|---|---|
| 1.0 | March 26, 2026 |
| 1.1 | April 30, 2026 |
Changes to objectives
| Version 1.0 | Version 1.1 |
|---|---|
| Objective 1.1.1: Define basic AI terms (for example, AI, ML, deep learning, neural networks, computer vision, natural language processing [NLP], model, algorithm, training and inferencing, bias, fairness, fit, large language models [LLMs]). | Objective 1.1.1: Define basic AI terms (for example, AI, ML, deep learning, neural networks, computer vision, natural language processing [NLP], model, algorithm, training and inferencing, bias, fairness, fit, large language model [LLM], generative AI [GenAI], agentic AI). |
| Objective 1.1.2: Describe the similarities and differences between AI, ML, GenAI, and deep learning. | Objective 1.1.2: Describe the similarities and differences between AI, ML, GenAI, deep learning, and agentic AI. |
| Objective 1.1.3: Describe various types of inferencing (for example, batch, real-time). | Objective 1.1.3: Describe various types of inferencing (for example, batch, real-time, asynchronous, serverless). |
| Objective 1.1.5: Describe supervised learning, unsupervised learning, and reinforcement learning. | Objective 1.1.5: Describe different types of AI/ML learning (for example, supervised learning, unsupervised learning, reinforcement learning methods). |
| Objective 1.2.4: Identify examples of real-world AI applications (for example, computer vision, NLP, speech recognition, recommendation systems, fraud detection, forecasting). | Objective 1.2.4: Identify examples of real-world AI applications (for example, computer vision, NLP, speech recognition, recommendation systems, fraud detection, forecasting, knowledge bases, agentic AI). |
| Objective 1.3.1: Describe components of an ML pipeline (for example, data collection, exploratory data analysis [EDA], data pre-processing, feature engineering, model training, hyperparameter tuning, evaluation, deployment, monitoring). | Objective 1.3.1: Describe and differentiate components of an AI/ML pipeline. |
| Objective 1.3.4: Identify relevant AWS services and features for each stage of an ML pipeline (for example, SageMaker AI, SageMaker Data Wrangler, SageMaker Feature Store, SageMaker Model Monitor). | Objective 1.3.4: Identify relevant AWS services and features for each stage of an AI/ML pipeline (for example, Amazon Bedrock, Amazon Q, Amazon Quick, Kiro, SageMaker AI). |
| Objective 1.3.6: Describe model performance metrics (for example, accuracy, Area Under the Curve [AUC], F1 score) and business metrics (for example, cost per user, development costs, customer feedback, return on investment [ROI]) to evaluate ML models. | Objective 1.3.6: Describe model performance metrics (for example, accuracy, precision, recall, F1 score) and business metrics (for example, cost per user, development costs, customer feedback, return on investment [ROI]) to evaluate ML models. |
| Objective 2.2.1: Describe the advantages of GenAI (for example, adaptability, responsiveness, simplicity). | Objective 2.2.1: Describe the advantages of GenAI (for example, adaptability, responsiveness, conversational capabilities, ability to generate content). |
| Objective 2.2.3: Identify factors to consider when selecting GenAI models (for example, model types, performance requirements, capabilities, constraints, compliance). | Objective 2.2.3: Identify factors to consider when selecting GenAI models (for example, model types, performance requirements, capabilities, constraints, compliance, cost, latency, model complexity). |
| Objective 2.2.4: Determine business value and metrics for GenAI applications (for example, cross-domain performance, efficiency, conversion rate, average revenue per user, accuracy, customer lifetime value). | Objective 2.2.4: Determine business value and metrics for GenAI applications (for example, cross-domain performance, ROI, efficiency, conversion rate, average revenue per user, accuracy, customer lifetime value). |
| Objective 2.3.1: Identify AWS services and features to develop GenAI applications (for example, Amazon SageMaker JumpStart, Amazon Bedrock PartyRock, Amazon Q, Amazon Bedrock Data Automation). | Objective 2.3.1: Identify AWS services and features to develop GenAI applications (for example, Amazon Bedrock, Amazon SageMaker AI, SageMaker JumpStart, Amazon Quick, Kiro, Strands Agents, Amazon Bedrock AgentCore). |
| Objective 3.1.5: Explain the cost tradeoffs of various approaches to FM customization (for example, pre-training, fine-tuning, in-context learning, RAG). | Objective 3.1.5: Explain the cost tradeoffs of various approaches to FM customization (for example, pre-training, fine-tuning, in-context learning, RAG, model distillation). |
| Objective 3.1.6: Describe the role of agents in multi-step tasks (for example, Amazon Bedrock Agents, agentic AI, model context protocol). | Objective 3.1.6: Define the role of AI agents and describe AI agents' business applications. |
| Objective 3.4.1: Determine approaches to evaluate FM performance (for example, human evaluation, benchmark datasets, Amazon Bedrock Model Evaluation). | Objective 3.4.1: Determine approaches to evaluate FM performance (for example, human-in-the-loop evaluation, benchmark datasets, Amazon Bedrock Model Evaluation). |
| Objective 3.4.2: Identify relevant metrics to assess FM performance (for example, Recall-Oriented Understudy for Gisting Evaluation [ROUGE], Bilingual Evaluation Understudy [BLEU], BERTScore). | Objective 3.4.2: Identify relevant metrics to assess FM performance (for example, Recall-Oriented Understudy for Gisting Evaluation [ROUGE], Bilingual Evaluation Understudy [BLEU], BERTScore, LLM-as-a-judge). |
| Objective 4.2.2: Describe tools to identify transparent and explainable models (for example, Amazon SageMaker Model Cards, open source models, data, licensing). | Objective 4.2.2: Describe tools to identify transparent and explainable models (for example, Amazon SageMaker Model Cards, SageMaker Clarify, Amazon Bedrock Model Evaluations, open source models, data, licensing). |
| Objective 4.2.4: Describe principles of human-centered design for explainable AI. | Objective 4.2.4: Describe principles of human-centered design for explainable AI (for example, user-feedback mechanisms, AI decision transparency). |
| Objective 5.1.1: Identify AWS services and features to secure AI systems (for example, IAM roles, policies, and permissions; encryption; Amazon Macie; AWS PrivateLink; AWS shared responsibility model). | Objective 5.1.1: Identify AWS services and features to secure AI systems (for example, IAM roles, policies, and permissions; encryption; Amazon Macie; AWS PrivateLink; AWS shared responsibility model; Amazon Bedrock AgentCore Identity; Policy in AgentCore; Amazon Bedrock Guardrails). |
| Objective 5.1.4: Describe security and privacy considerations for AI systems (for example, application security, threat detection, vulnerability management, infrastructure protection, prompt injection, encryption at rest and in transit). | Objective 5.1.4: Describe security and privacy considerations for AI systems (for example, application security, threat detection, vulnerability management, infrastructure protection, prompt injection, encryption at rest and in transit, data leakage prevention, output filtering and validation, audit trail and logging requirements for AI interactions, toxicity). |
Objectives added
Objective 1.2.6: Identify when traditional ML models versus foundation models (FMs) are appropriate for a given use case (for example, due to regulatory concerns, explainability, operational constraints).
Objective 2.1.4: Describe the token-based pricing model and its effect on cost and performance for inference.
Objective 2.1.5: Describe the role of context engineering in FM applications.
Objective 2.1.6: Define foundational agentic AI concepts (for example, multi-agent system patterns for complex AI applications, Model Context Protocol [MCP] and its role in connecting agents to external systems, multi-agent communication patterns, memory management, tool usage, and workflow orchestration).
Objective 3.2.5: Describe prompt versioning and management strategies that use Amazon Bedrock Prompt Management.
Objective 3.4.5: Identify business objective alignment metrics for AI applications (for example, task completion rate, user satisfaction, cost per interaction).
Objective 5.1.5: Describe hallucination detection methods and grounding techniques to improve output accuracy (for example, Retrieval Augmented Generation [RAG] grounding, output validation, confidence scoring).
Changes to in- and out-of-scope services
Services added to the in-scope list
Amazon Aurora
Amazon Bedrock AgentCore
Kiro
Strands Agents
Amazon Q
Amazon SageMaker JumpStart
AWS Transform
Services removed from the in-scope list
Amazon MemoryDB
Services removed from the out-of-scope list
AWS DeepComposer
Amazon FinSpace
Amazon Honeycode
AWS IAM Identity Center
AWS Marketplace
AWS Organizations
Amazon WorkDocs
