Agentic AI for Dummies.

Format:

Book

Author/Creator:

Baker, Pam.

Language:

English

Subjects (All):

Artificial intelligence.

Physical Description:

1 online resource (354 pages)

Edition:

1st ed.

Place of Publication:

Newark : John Wiley & Sons, Incorporated, 2026.

Summary:

An easy-to-follow guide to demystifying Agentic AI, the next step in the evolution of artificial intelligence Agentic AI is the next big leap in artificial intelligence.Agentic systems don't just respond to commands.They set goals, make decisions, and take initiative without direct human interaction.Sound like a lot to wrap your head around?.

Contents:

Intro

Title Page

Copyright Page

Table of Contents

Introduction

About This Book

Foolish Assumptions

Icons Used in This Book

Beyond the Book

Where to Go from Here

Part 1 Understanding Agentic AI

Chapter 1 Introducing Agentic AI

Defining Agentic AI

Moving toward AGI with Agentic AI

Noting that agentic systems already exist

Reasoning as AI's Way Forward

Exploring philosophy, reason, and fear

Putting AI reasoning into perspective

Recognizing the operational challenges of Agentic AI

Differentiating between AI Agents and Agentic AI

Meeting examples of agents and agentic systems

Seeing what agents and Agentic AI systems have in common

Recognizing the distinctions between agents and Agentic AI systems

Mapping the Path from Prompt Engineering to AI Autonomy

Engineering your prompts for successful AI interaction

Controlling AI system actions with effective prompting

Viewing prompting as foundational

The Budding Agentic AI Web

Expanding the duty of agents online

Scaling up to a citywide, nationwide, and global reach

Following the Shift to A-Commerce

Recognizing the fallibility of AI answers

Reducing specific website interaction

Relying on personal AI shoppers

Optimizing commerce sites for AI

Navigating the transition to A-commerce

Chapter 2 Peeking Inside the AI Agent Mind

Linking the Fundamental Building Blocks

Identifying Agentic AI building blocks

Enter the overseers

Exploring Reasoning, Memory, and Goal Setting

Assessing Agentic AI reasoning

Considering AI's limited intelligence

Recognizing its intentional design

Evaluating Agentic AI memory

Adding memory to a system's design

Blending memory and reasoning in the design

Getting (too) personal with memory

Grasping Agentic AI goal setting.

Understanding Adaptive Behavior and Self-Directed Learning

Dissecting adaptive behavior

Delving into self-directed learning

Learning more than new information

Examining other aspects of meta-learning

Directing Agentic AI

Talking it over with Agentic AI

Continuing direction over the AI's work

Completing the mission and next steps

Interacting with GenAI and Agentic AI

Combining Generative Abilities and Real-Time Decision-Making

Expanding on content generation

Applying agentic capabilities to complex interconnections

Operating autonomously across time

Staying the course in a changing environment

Chapter 3 Meeting Agentic AI Core Technologies

Driving Multi-Agent Coordination and Planning

Computational complexity and task decomposition

Specialized expertise and division of labor

Scalability and fault tolerance

Distributed information and resources

When resources can't be moved

Apply the multi-agent approach

Emerging coordination mechanisms

Communication and shared understanding

Connecting Contextual Awareness and Situational Reasoning

World modeling

Perception and sensor fusion

Memory architectures

Theory of mind modeling

Communication protocols and intent signaling

Planning and goal-conditioned learning

Distributed coordination and federated learning

Self-Correcting Continuous Improvement

Improving by failing and adjusting

Correcting more than just mistakes

Shifting to Multimodal Input and Cross-Domain Functionality

Contrasting reactive and proactive operation

Exploiting multiple streams of input

Enabling the growth of multimodal systems

Streamlining Integrations Using New Protocols

Model Context Protocol (MCP)

Seeing how MCP works

Recognizing MCP's limitations

Agent Network Protocol

Agent2Agent protocol (A2A).

Fostering agent communication

Barriers to agent collaboration

Agent Communication Protocol

Incorporating internet methods

Focusing on precision

Building AI Agents

Choosing technical architecture approaches

Building from scratch

Using agentic frameworks

Using AI agent-building platforms

Supporting system development, regardless of method

Building without coding

Recognizing both benefits and drawbacks

Looking over types of platforms

Building with coding

From-scratch options (or not)

Framework options

Chapter 4 Interacting with Agentic AI

Mistaking AI as a Colleague Creates Errors

Establishing the AI-as-tool mindset

Discovering how to direct Agentic AI

Comparing Context Engineering to Prompt Engineering

Seeing why you need both practices

Understanding the fundamental differences in engineering methods

Examining the basics of context engineering

Providing comprehensive awareness

Interconnecting system components

Augmenting context engineering with prompt engineering

Incorporating prompt engineering in Agentic AI

Prompting and tool integration

Maintaining agent behavior

Evolving Voice, Intent, and Semantic Interface Design

Voicing your intent with AI

Interpreting meaning with semantic interfaces

Rising Hyper-Real AI Avatars

Personalizing Workflows

Taking informed actions

Adapting to (and for) user's roles

Tailoring output based on user feedback or co-agent needs

Shifting from Apps to Agents

Reducing the complexity of our world

Dying app stores

Reimagining the internet through agents

Grappling with AI-related challenges

Forbidding AI Agents from Running Certain Machines

Recognizing the system design differences

Addressing the timing issue

Noting the current lack of Agentic AI transparency.

Exposing the interface issues

Examining data integrity

Part 2 Getting Started on the Agentic AI Path

Chapter 5 Planning for the Shift to Agentic AI Systems

Comparing Generative AI to Agentic AI with Goals in Mind

Considering AI strengths and oversight required

Seeing the power of combined AI

Thinking Through an Agentic AI Plan

Recognizing the five pillars of Agentic AI planning

Setting up SMART goals and detailed follow-up

Double-checking your plan

Following the Steps for Planning and Implementing Agentic AI

Step 1: Establishing strategic intent

Making a strategic-intent commitment

Examining the components of strategic intent

Step 2: Evaluating readiness

Step 3: Identifying high-impact use cases

Linking Agentic AI deployments to business priorities

Looking for value and success

Choosing a pilot project

Step 4: Designing the pilot framework

Step 5: Building or integrating Agentic AI systems

Implementing technical strategies

Integration architecture and data flow

Monitoring and observability systems

Deployment and operational considerations

Step 6: Running, measuring, and refining

Taking the right measurements

Refining operational system models

Step 7: Expand and scale

Facing challenges of Agentic AI expansion

Starting small and scaling up

Meeting organizational complexity

Step 8: Establishing governance and trust

Step 9: Upskilling your workforce

Step 10: Reimagining business models and value creation

Chapter 6 Sampling Sector Use Cases for AI Agents

Developing Healthcare, Diagnostics, and Pharmaceuticals

Personalizing with AI treatment agents

Dispatching AI medical imaging agents

Accelerating clinical trial patient matching

Upskilling surgeons for robotic surgery

Building Business Operations and Decision Support.

Simulating economies with AI financial modeling agents

Optimizing procurement and negotiations with AI agents

Auditing with ever-present internal AI agents

Adding AI Agents for Marketing, Customer Experience, and Inventory

Deploying AI marketing agents

Engaging AI customer experience agents

Building AI inventory optimization agents

Creating Content: Writing, Design, and Media

Collaborating with AI writing agents

Designing with AI agents

Editing with post-production agents

Reinventing Education

Personalizing AI learning agents

Contextually aware tutoring agents

Upgrading administration automation

Chapter 7 Considering Risks, Ethics, and Hard Questions

Losing Human Skill and Baseline Knowledge

Seeing the scope of AI-involved skill loss

Delegating work to AI agents responsibly

Designing agentic systems to mitigate loss of competency

Autonomy versus Control: Establishing Who's in Charge

Transportation: Lessons in control

Healthcare: Human oversight in life-and-death decisions

Finance: Algorithms, autonomy, and accountability

Education: Autonomy and human agency

Discovering Alignment Problems and Value Misfires

Value learning drift

Everyday misfires

High-stakes alignment challenges

Why achieving alignment is so hard

Detecting misfires

Value alignment as a collective effort

Missing Transparency and Explainability

Looking for transparency in Agentic AI reasoning

Addressing explainability

Revisiting Bias, Justice, and Inclusivity

From hidden bias to active misfires

Justice as more than accuracy

Inclusivity as a design imperative

Cultural and contextual sensitivity in autonomous operations

Hallucinating AI Agents at the Wheel?

Addressing AI hallucinations

Aiding AI with clear direction and human oversight.

Part 3 Agentic AI in the Real World.

Notes:

Description based on publisher supplied metadata and other sources.

ISBN:

1-394-37962-5

OCLC:

1561171904