Accelerating AI innovation:

A comprehensive guide
for technology leaders
Discover how to build a foundation for AI success in the agentic era
Table of contents

Introduction: Are you ready to capitalize on AI? 3 AI you can trust 12

Drive real business value with gen AI 4 A commitment to safe, secure, and transparent
13
AI services and tools
AI brings opportunities and challenges 5 14

Tools for building responsibly with AI
How to make AI implementation easier and better 6 Maximizing value with AI 15
Innovate without constraints: Build a foundation
for AI success
7 AI use cases for every business 16
Applications to boost productivity 8 Customer success stories 17

AI and agent development software and services 9 Next steps: Fuel gen AI breakthroughs with AWS
21
Infrastructure to build and train AI models 10
Make data your differentiator 11

2
 INTRODUCTION

Are you ready to 80%

capitalize on AI?
AI is delivering real business value today, helping organizations reimagine customer experiences,
boost productivity, and drive innovation. And now, with the emergence of agentic AI, organizations
are exploring how to leverage AI systems that can autonomously perform tasks and make decisions to
More than 80% of enterprises will have
further enhance their operations. deployed gen AI applications by 2026¹

To realize this future, organizations need more than a single, powerful large language model (LLM) or
chat assistant. They need a full range of capabilities to build and scale AI applications and agents that
are tailored to their business and use cases—including applications with built-in gen AI, tools to rapidly
build their own apps and agents, a cost-effective and high-performance infrastructure, and robust
security controls.

While AI is accelerating, many organizations face challenges moving from proofs of concept to real-
world applications. Another universal challenge is identifying and implementing the right use cases to 30%
align with specific business goals. This guide is designed to help you unlock the value of AI quickly, scale
securely, and innovate with higher performance and lower cost.

At least 30% of gen AI projects will be

abandoned after proof of concept by the
end of 2025 due to poor data quality,
inadequate risk controls, escalating
costs, or unclear business value2

¹ “Gartner Says More Than 80% of Enterprises Will Have Used Generative AI APIs or Deployed Generative AI-Enabled Applications by 2026,” Gartner, October 2023
² “Gartner Predicts 30% of Generative AI Projects Will Be Abandoned After Proof of Concept By End of 2025,” Gartner, July 2024 3
 BUSINESS OPPORTUNITIES

Gen AI has the

Drive real business value potential to create

with gen AI
Gen AI has the potential to create $2.6 trillion–$4.4 trillion in business value globally.³ It can help your
$2.6T–$4.4T
organization in four fundamental ways:
in business value globally³

Creating new experiences Extracting insights

Deliver innovative and engaging ways to engage Surface valuable insights from large volumes of
with customers and employees. documents and data to improve decision making.

Boosting employee Enhancing creativity

productivity Generate new content, ideas, conversations,
and multimedia.
Empower employees to create content, code,
and ideas more efficiently.

3
“The economic potential of generative AI: The next productivity frontier,” McKinsey & Company, June 2023
4
 BUSINESS CONSIDERATIONS
Key challenges in implementing AI
AI brings Data infrastructure modernization, integration,

opportunities and scalability

Legacy systems inhibit advanced analytics and AI capabilities and bring

and challenges substantial capacity constraints.

Regulatory, ethical, data sovereignty, and data

As a technology leader, you play a pivotal role in driving the successful adoption residency considerations
of gen AI and agents in your organization. However, you must navigate the
complexities of this cutting-edge technology and strike a balance between seizing Leaders must navigate the complex, uncertain, and increasingly ambiguous
the opportunities and mitigating the risks. regulatory landscapes and comply with relevant laws.

Balancing costs while maintaining performance

Training, building, and deploying gen AI models is challenging with budgets
that remain close to flat year over year.

Model selection
As the landscape of language models continues to evolve, technical leaders
are challenged to identify the most suitable model for each use case.

5
How to make AI implementation
easier and better
In helping hundreds of thousands of customers move from proof of concept to production, we’ve found that successful AI
implementation comes down to three key elements: the freedom to innovate without constraints, AI you can trust, and the ability
to maximize value at scale. This is why Amazon Web Services (AWS) delivers AI innovation that meets you wherever you are on your
journey through a comprehensive and flexible AI portfolio backed by unmatched scale and expertise.

Giving you the Providing you with Helping you

freedom to invent AI you can trust maximize value

6
 FREEDOM TO INVENT
AWS AI stack
Innovate without Applications to boost productivity
constraints: Build
a foundation for AI and agent development software and services
AI success
As you move from opportunity to implementation, your organization
will need these key elements to succeed with AI:
Infrastructure to build and train AI models

1. A flexible foundation that gives you a choice of tools and models.

AI is evolving rapidly, and your infrastructure needs to evolve
with it.

2x
2. A strong data strategy. Your organization’s unique data is what
will ultimately differentiate your AI solutions and drive real AWS invests at every
business value.
layer of the AI stack,
3. Agentic AI capabilities. Building intelligent systems capable of
handling complex workflows autonomously will become essential
from AI assistants to
more production-ready capabilities
for maintaining competitive advantage in the agentic era. foundation models (FMs) than other cloud providers to
to chips, and provides: support the path you choose

7
 AWS AI STACK SPOTLIGHT

Meet the most capable AI agent assistants

Applications to boost AWS Transform is the first agentic AI service developed to accelerate enterprise

productivity modernization of .NET, mainframe, and VMware workloads. Built on 19 years of

migration experience, it deploys specialized AI agents to automate complex tasks
like code analysis, refactoring, decomposition, dependency mapping, validation,
Accelerate your AI journey with ready-to-deploy solutions. AWS offers powerful and transformation planning, dramatically reducing project timelines.
applications and agents that can help you take advantage of the productivity
Amazon Quick Suite transforms workplace productivity by empowering every
benefits of AI quickly.
business user with AI agents that streamline complex tasks, helping them make
better decisions, faster, and act on them swiftly. By seamlessly integrating
comprehensive research capabilities, data-driven business insights, and powerful
agentic automation tools into a single experience, Amazon Quick Suite helps

30%
users work smarter—all while maintaining security and user access policies.

Kiro is an AI integrated development environment (IDE) that helps you go from

prototype to production with spec-driven development. From simple to complex
tasks, Kiro works alongside you to turn prompts into detailed specs, then into
AWS Transform helped working code, docs, and tests—so what you build is exactly what you want and
Thompson Reuters cut costs ready to share with your team. Kiro’s agents help you solve challenging problems
and automate tasks like generating documentation and unit tests. With Kiro, you
by 30% and boost .NET
can build beyond prototypes while being in the driver’s seat every step of the way.
modernization speed by 4x

8
 AWS AI STACK SPOTLIGHT

AI and agent development Build effective and responsible AI apps

and agents with Amazon Bedrock
software and services Amazon Bedrock is a fully managed service that gives you access to leading FMs
and tools to build secure, customizable gen AI applications and agents with the
For customers seeking access to the latest models, flexibility to select the performance and reliability your business demands.
appropriate model for their use case, and robust tools and security controls to
swiftly build and scale trustworthy AI apps and agents, we offer Amazon Bedrock Amazon Bedrock AgentCore enables you to deploy and operate highly capable AI
and Amazon Bedrock AgentCore. agents securely, at scale. It offers infrastructure purpose-built for dynamic agent
workloads, powerful tools to enhance agents, and essential controls for real-
world deployment. Amazon Bedrock AgentCore services can be used together or
independently and work with any framework.

100+
AI MODELS Optimize your AI apps with powerful,
cost-effective models
Available through
Amazon Bedrock Amazon Nova foundation models deliver frontier intelligence and industry-
leading price performance. Building with Amazon Nova models in Amazon
Bedrock, organizations can build and scale AI applications that are safe, reliable,
and cost-effective.

9
 AWS AI STACK SPOTLIGHT

Infrastructure to build and The most performant, cost-effective

infrastructure for gen AI
train AI models From the highest-performance NVIDIA GPU–based Amazon Elastic Compute
Cloud (Amazon EC2) instances to our purpose-built machine learning (ML)
To cost-effectively build, train, and deploy FMs at scale, you need purpose-built accelerators AWS Trainium and AWS Inferentia, AWS delivers the best price
infrastructure and services. performance for training and deploying gen AI models at scale.

With Amazon SageMaker AI, data scientists and ML engineers can easily build,
train, and deploy FMs at scale with purpose-built tools, including IDEs and
notebooks backed by high performance accelerated computing, purpose-built
infrastructure for distributed training at scale, governance and machine learning

40%
operations (MLOps) tools, inference options and recommendations, and model
monitoring and evaluation.

Amazon SageMaker HyperPod reduces

training time by up to 40% on AWS

10
 UNLOCK THE VALUE OF YOUR DATA

Make data your

differentiator 52% of CDOs and data
leaders say they
don’t have the right
data foundation to
Your data is the differentiator between a generic application and a gen AI
application that knows your business and your customer. AWS makes it
implement gen AI⁴
easy to use your data to customize and augment FMs to offer personalized
experiences to your customers. AWS also offers a comprehensive range of data
services to store, process, manage, and govern your data. The next generation
of Amazon SageMaker addresses the challenges of harnessing all of your
organizational data—regardless of where it lives—for analytics and AI through
unified data access and governance. It enables teams to securely find, prepare,
and collaborate on data assets and build analytics and gen AI applications
through a single platform, accelerating the path to AI-ready data.

Discover how a data foundation built on AWS gives you a strategic advantage
when it comes to gen AI.

⁴ “Scaling Generative AI for Value: Data Leader Agenda for 2025,” Harvard Business Review Research Report, sponsored by AWS, 2024
11
AI you can trust AMAZON BEDROCK

Security from the start

Building AI you can trust isn’t just an AWS promise—it’s a fundamental
requirement for your business success. This trust is built on the following:
Customer data is never shared with FM providers

Customer data is always encrypted with a

1
minimum of TLS 1.2
Robust security and privacy protections to safeguard your
data and intellectual property Monitor API activity and also troubleshoot
issues with AWS CloudTrail

2 Reliable and accurate outputs that provide a consistent

experience for the users of your AI applications

12
 RESPONSIBLE AI

A commitment to safe, Responsible AI

secure, and transparent AI Responsible AI is the practice of designing, developing, and

services and tools using AI technology in a responsible manner, with the goal
of maximizing benefits and minimizing risks. At AWS, we
define responsible AI using a set of core dimensions that we
You want your gen AI applications to be built with safety, fairness, and security assess and update over time as AI technology evolves.
in mind so that you can, in turn, deploy AI responsibly. Our practical approach
to transforming responsible AI from theory into practice, coupled with tools and
expertise, can help you implement responsible AI practices effectively within
your organization. We are investing in new capabilities to foster responsible gen
AI innovation across a broad set of capabilities, including built-in tools, to help
promote fairness, transparency, and customer protection, as well as resources to
combat disinformation.
Core dimensions of responsible AI
Learn more about responsible AI
Controllability Veracity and
robustness
Privacy and
security Explainability

Safety Transparency

Fairness Governance

13
 RESPONSIBLE AI

Tools for building

responsibly with AI
• Amazon Bedrock Guardrails helps users implement safeguards tailored to gen
AI applications and aligned with responsible AI policies.
• Amazon Bedrock Evaluations help users evaluate, compare, and select the
best FMs for their specific use case based on custom metrics, such as accuracy,
robustness, and toxicity.
• Amazon Nova Reel and Amazon Nova Canvas have built-in controls for the
safe and responsible use of AI. These include watermarking for traceability,
content moderation, and indemnification.
• AWS AI Service Cards are a resource for enhancing transparency by providing
users with a single place to find information on use cases and limitations,
responsible AI design choices, and performance optimization best practices
for our AI services and models.

14
Maximizing value with AI Amazon Bedrock offers several powerful optimization
tools, such as model distillation, intelligent prompt
One of the most critical aspects of AI adoption is maximizing business value while routing, prompt caching, and flexible payment
managing costs.
options, that can improve latency by up to
500 percent and reduce costs by up to 75 percent
Success requires two key elements:
while sacrificing less than 2 percent in accuracy
• Being strategic about model selection and customization to improve performance
and drive return on investment (ROI)
• Focusing on high-impact use cases that deliver tangible business outcomes

AWS provides tools that help you pursue ambitious AI initiatives while maintaining
strong cost control, ensuring your investments translate into measurable results.

15
 USE CASES Some of the high-impact areas where
gen AI can drive value include:
AI use cases for every
business Enhancing customer experiences
• Chatbots and virtual assistants
Gen AI has the potential to transform nearly every industry and function
within your organization. The key is identifying the right use cases that • Agent assist and call analytics
align with your specific business needs and goals. Ask yourself: What • Hyper-personalization
requirement is most critical to the organization’s success?

Boosting employee productivity

• Conversational search
• Code generation
• Automated report generation
• AI-generated marketing content, sales content,
guidance, and enablement
• New product development

Optimizing business processes

• Intelligent document processing (IDP)
• Data augmentation
• Supply chain optimization

16
 CUSTOMER SUCCESS STORIES

Customers across industries are

enhancing customer experiences,
boosting employee productivity, and
optimizing business processes with
gen AI apps and agents on AWS.
FERRARI

3M

AMAZON ALEXA

17
Ferrari redefines automotive luxury
Ferrari, a luxury Italian auto manufacturer, is using the broad model
selection of Amazon Bedrock to apply gen AI to several use cases, from
accelerating the vehicle design process to providing personalized services
to its customers. Using LLMs in Amazon Bedrock, Ferrari developed a car
configurator to make it easier and faster for customers to personalize
their car, which increased sales leads and reduced vehicle configuration
times by 20 percent. Ferrari also fine-tuned LLMs—including Amazon
Titan, Claude 3, and Llama—on its own internal documentation to create
a gen AI chatbot that helped its sales professionals and technicians
enhance the after-sales experience.

“Amazon Bedrock has simplified our approach. We

can connect to a single layer of APIs to quickly test,
benchmark, and deploy different models.”
Mauro Coletto, Head of Business Analytics & AI at Ferrari

18
3M leverages the capabilities of
Amazon Quick Suite
At 3M, sales teams conduct numerous client meetings and spend
excessive time piecing together fragmented information scattered
across multiple systems. By leveraging Amazon Quick Suite,
an AI-powered workspace, sales teams can quickly get insights
through natural conversations with AI agents and automate routine
tasks using Quick Flows. This approach is helping drive pipeline
progression and opportunities. Additionally, other teams have
found value with Quick Research, compressing weeks of intensive
analysis into expert-level insights delivered the same day.

19
Amazon transforms Alexa into a personal
AI assistant
Amazon leveraged Amazon Bedrock and Amazon Nova to create
Alexa+—a next-generation AI assistant that combines conversational
abilities with autonomous task execution. The solution integrates state-
of-the-art language models with practical task automation capabilities
to handle complex requests and proactively manage daily activities.
Built on the secure infrastructure of AWS, Alexa+ brings world-class
privacy and security protection to your everyday interactions.

“Rebuilt from the ground up, this new version of Alexa uses
a state-of-the-art architecture that automatically connects a
variety of large language models (LLMs), agentic capabilities,
services, and devices at scale.”
Amazon Staff

20
 NEXT STEPS AWS expertise

Fuel gen AI breakthroughs

Take advantage of AWS experts and partners to
innovate faster with gen AI.

with AWS AWS Generative AI

Innovation Center
A program pairing customers with AWS science
The future of business is being shaped by AI, and AWS meets you wherever you are on your journey.
and strategy experts with comprehensive
Whether you want to build custom AI solutions, deploy pre-built applications, or partner with experts,
expertise spanning the entire gen AI journey.
AWS provides the flexibility to choose your path while maintaining control of your data and costs.
The AWS Generative AI Innovation Center helps
Join the hundreds of thousands of AWS customers using our secure, flexible, and scalable AI services customers build a strategic and holistic approach
to innovate continuously, deliver exceptional customer experiences, and drive sustained growth—all aligned with business goals, use cases, and
while maintaining the highest levels of trust, privacy, and security. operational challenges.

Start now AWS Generative AI

Competency Partners
AWS Generative AI Competency Partners drive
the advancement of gen AI technologies with
AWS customers to launch transformative
applications across industries.

© 2025, Amazon Web Services, Inc. or its affiliates. All rights reserved.
21