The State of

AI Infrastructure
2024 Edition

An annual report on trends and developments

in AI infrastructure based on Microsoft-
commissioned surveys conducted by
Forrester Consulting and Ipsos.
AI is sweeping across industries and transforming the
way businesses operate, innovate, and compete. But how are
organizations actually using AI today, and what are their plans
for the future? What are the main challenges and priorities
they face in implementing AI at scale? And what role does
infrastructure play in enabling AI adoption and performance?

We explore these questions and more, based on surveys of

over 1,500 business leaders from various sectors and regions.
We also share insights and best practices from leading-edge
organizations that are already leveraging AI to create value and
drive growth. Whether you are just starting your AI journey or
looking to scale it up, this paper will help you understand the
state of AI and AI infrastructure, and how to build a strategy
that works for your business.

2
Table of contents

AI is here. It’s just the beginning__________________________________ 4

AI is challenging…for everyone ___________________________________ 9

AI infrastructure remains elusive_________________________________ 13

Start with the trifecta: performance, security, and cost _________ 17

One size does not fit all __________________________________________20

Harnessing the power of AI now _________________________________25

Take the next steps on your AI journey __________________________27

Research methodology___________________________________________28

3
AI is here.
It’s just the beginning.
Welcome to the new era, where AI is not only intriguing and engaging consumers
but exponentially increasing business productivity, transforming business models,
and reimagining customer experiences. From retail to healthcare, there’s no doubt
AI is making a difference in ways like:

• Aggregating customer data to serve personalized recommendations at scale in

retail.

• Powering CT scanners with robust algorithms for more accurate diagnostics and
improved medical care.

• Predicting machine lifecycles for real-time maintenance and run-time

efficiencies in factories.

• Preventing financial fraud via real-time detection using advanced AI tools and
models.

• Using consumer-friendly chatbots to streamline customer service processes.

Recent Microsoft-commissioned research shows most companies are actively

ramping up their AI capabilities, with 95% of businesses surveyed planning to
increase their AI usage over the next two years. Across industries, AI adoption is
believed to be critical for success.

Collectively, there is a consensus over the importance of AI, not just from an
organizational standpoint, but also from a personal standpoint. This is an important
distinction – these numbers show that it’s not only organizations that are driving
adoption, but that people see the value personally.

Source: A commissioned study conducted by Forrester Consulting on behalf of Microsoft and NVIDIA, May 2023

4
AI importance for success

AI is critical to my organization’s success

Strongly agree/Agree

Total 75%

Finance 73%

Healthcare 72%

Retail 76%

Manufacturing 80%

ISV 75%

AI is critical to my personal success Strongly agree/Agree

Total 66%

Finance 68%

Healthcare 62%

Retail 66%

Manufacturing 73%

ISV 64%

Base: Total (n=900), Finance (n=180), Healthcare (n=180), Retail (n=180), Manufacturing (n=180), ISV (n=180)
Source: A commissioned study conducted by Ipsos on behalf of Microsoft, October 2023

5
Many organizations are at the starting line
More than a third of companies are in the early stages of AI adoption: exploring options,
gathering information, and planning various use-cases to strategize implementation,
while a quarter are in the early pilot testing stage. With a majority of organizations still
figuring things out, business leaders have an opportunity to beat their competition and
gain advantages. But to do so, they’ll need to act quickly in implementing their own AI
strategies.

Organizations are in different stages of AI implementation

Organizations in this earliest stage

are in the initial stages of exploring

38%
options, gathering information, and
identifying AI use cases as well as
planning and strategizing their AI
implementation. Organizations in this final stage
have fully implemented and
are both actively utilizing and
optimizing.

Organizations in the pilot stage

are currently conducting small-
scale implementations to evaluate
15%
25%
feasibility and effectiveness.

22% Deployment means that an organization

has deployed in specific areas and is
gradually expanding usage.

Base: Total (n=900), US (n=500), Germany (n=200), India (n=200).

Source: A commissioned study conducted by Ipsos on behalf of Microsoft, October 2023

6
Businesses are focused on automation
and customers
Those companies that have started integrating AI are focused on supporting customer-
facing applications and increasing efficiency through automation. These use cases
tend to bring higher ROI, as they focus on getting more value out of their workers by
reducing time spent on lower-value tasks. This makes sense, considering respondents,
on average, expect a 34% ROI from their AI platforms.

On average, 46% of customer-facing applications and 44% of

business/core applications leverage AI functionality.

70% 62%
Conversational AI (e.g.,
Document-process
customer service assistant,
automation (e.g., claims
language understanding,
management and automation)

59%
voice control)

Machine translation (e.g., real-

time speech translation, document
translation, web localization)​

52%
Speech transcription and
analytics (e.g., call transcription
and analysis, multimedia content
What AI workloads captioning, conversation
transcription)​
has your organization
deployed? 49%
Knowledge mining (e.g., content
search, product discovery
optimization)

Base: (n=641)
Source: A commissioned study conducted by Forrester Consulting on behalf of Microsoft and NVIDIA, May 2023

7
“Leading-edge organizations” blaze the
path forward
While many organizations are early in their AI journeys, 15% of businesses are advanced
in their AI infrastructure and are considered “leading-edge organizations”. These
leading-edge organizations tend to be early adopters of technology and can provide
valuable learnings on effective AI implementation strategies. Our recommendations are
based on an analysis of leading-edge organizations plus other key insights to provide AI
best practices and recommendations any company can leverage.

Stages of AI readiness

38% AI beginner 15% Leading-edge

Exploring and planning Fully implemented and
optimizing

71% of leaders in leading-

edge companies say that their
organization is eager to try new
technology and having IT that’s
more advanced than many of
their competitors compared
to 47% of non-leading-edge
47% Developing companies.
Piloting, testing, and deploying

Base: Total (n=900)

Source: A commissioned study conducted by Ipsos on behalf of Microsoft, October 2023

8
AI is challenging…
for everyone
The AI landscape continues to evolve and AI implementation poses many different
challenges and obstacles to overcome. Business leaders are faced with the daunting
task of figuring out the best path forward.

99%
99% 59%
59%

of organizations have of business leaders

challenges in scaling and believe the AI market is
operationalizing AI growing and evolving

My technology environment is very complex and dynamically changing.

We have complex needs and multiple departments use different applications...

all with varying needs.

In just a few years, I believe the AI will be much more advanced than what
we have today.

Base: Total (n=900)

Source: A commissioned study conducted by Ipsos on behalf of Microsoft, October 2023

9
The greatest challenge? Tackling the
talent gap
Organizations have an immediate need for AI experience and talent. Addressing this
gap by bolstering employees’ skills and training today is the key to bridging the gap
and moving forward quickly.

39% 33%

Leaders who rank having the skills Leaders who rank having enough
required to develop or customize talent as one of their
AI models as one of their top 3 top 3 organizational challenges
technology challenges (out of 13 items). (out of 13 items).

Security, capabilities, and ROI

considerations
Along with AI talent sourcing, it’s not surprising that many of the other challenges
businesses face are centered around technological and strategic challenges. Security
considerations, having adequate capabilities for designing, implementing, and
managing infrastructure, and having the appropriate AI tools are key technology
challenges.

And along with employee talent, unclear ROI of AI implementation, the right resources
to support AI development and management and collaborating across business
functions are top organizational challenges.

Base: Total (n=900)

Source: A commissioned study conducted by Ipsos on behalf of Microsoft, October 2023

10
Infrastructure challenges remain top of mind
The right infrastructure can make or break your AI projects. Overwhelmingly, the AI
challenges fall within infrastructure (hardware, software, and tools) and remain the most
common roadblock in implementing and leveraging powerful AI tools. Prioritizing the
right AI infrastructure is key to successful AI implementation, scaling, and innovation.

My organization doesn't have the proper

56% infrastructure to support my organization's desired

AI workloads.

Top technology chal

challenges organizations face
(descending order, showing top 7 of 13 items)

Skills required to develop or customize AI models

Security considerations Related to

infrastructure
H
 aving adequate capabilities for setting up, scaling,
and managing AI infrastructure

Accessing appropriate AI methods/tools

Outdated/legacy systems

Ability to scale infrastructure on demand

Orchestrating workloads across cloud and on-premises

Base: Total (n=641), Total (n=900)

Source: Commissioned studies conducted by Forrester Consulting and Ipsos on behalf of Microsoft, May 2023 and October 2023

11
Leaders are looking to partners for help
For companies not sure how to start leveraging AI, partnering with a solution provider
with deep AI expertise and proven AI solutions can help companies accelerate AI
production and address AI infrastructure challenges. Business leaders are looking to
partners to help with infrastructure design and implementation, training and support,
security and compliance, and strategic planning and consultation.

Where it gets interesting is that as companies move further along their AI journey, they
start to prioritize things like performance, optimization, and cloud provider integration.
Engaging the right partner can help businesses of any size and at any stage of AI
implementation accelerate their AI journey. This is both a huge opportunity for partners,
and a burden. They must make sure their staff is ready to go and able to help with
consulting, strategy, and training.

Where partners are expected to help (top 4)

Leading-edge Non-leading-edge

Performance
Training and
monitoring and 47% 47% support
52% 52%
optimization

Security and Security and

44% 44% 50% 50%
compliance compliance

Infrastructure Infrastructure
design and 44% 44% design and 47% 47%
implementation implementation

Cloud provider Strategic planning

44% 44% 44% 44%
integration and consulting

Base: Total (n=900)

Source: A commissioned study conducted by Ipsos on behalf of Microsoft, October 2023

12
AI infrastructure remains
elusive
If you’re not sure how to approach your AI infrastructure, you’re not alone. Robust and
scalable infrastructure specifically built for AI is critical to support the complexities of
new AI-driven workloads and processes, but AI infrastructure is a key challenge for most
leaders, who face many obstacles in implementing and operationalizing AI,
such as:

• Outdated and legacy systems that are not designed to handle the complexity and
volume of AI workloads.

• Data security and privacy concerns, especially sensitive and personal data,
that require robust protection and compliance measures.

• Workload orchestration challenges, such as managing multiple platforms,

tools, and frameworks, and optimizing resource utilization and performance.

• Skills gap, as many organizations lack the talent and expertise to develop, customize,
and deploy AI models and applications.

• The accelerated rate of technological advancements like GenAI that have

large implications on the type and complexity of the infrastructure needed.

13
Defining AI infrastructure
Infrastructure challenges are amplified due to diverse interpretations of AI
infrastructure among organizations. These interpretations range from integrating AI
capabilities into the existing IT infrastructure to establishing a dedicated hardware
and network infrastructure and developing a comprehensive tech stack that includes
algorithms, frameworks, and libraries. This can make something as simple as
communicating needs with vendors a real challenge when the same language is not
being used.

Not having a clear definition just adds to the challenges of getting started with AI. At
the most simplistic, AI infrastructure includes the hardware, software, networking, and
tools and services used to develop, implement, and optimize AI. As AI continues to
evolve, it’ll be more important than ever to settle on a standard definition that can be
used across all industries.

The different ways organizations define “AI infrastructure”

Integrating AI capabilities into existing IT
infrastructure through the utilization of cloud 57%
resources, services, and APIs

A combination of hardware, software, and tools

that enable the development, deployment, and 54%
management of AI models and applications

A comprehensive set of AI-specific algorithms,

frameworks, and libraries that form the foundation 39%
of the company’s AI capabilities

Dedicated hardware and network infrastructure

specifically designed to support AI initiatives 34%
(e.g., high-performance servers or GPUs,
reliable network infrastructure, etc.)

Microsoft’s definition of AI infrastructure: “A combination of

hardware, software, and tools that enable the development,
deployment, and management of AI models and applications.”

Base: Total (n=900)

Source: A commissioned study conducted by Ipsos on behalf of Microsoft, October 2023

14
Making AI infrastructure a priority
Businesses are already understanding the urgency and importance of having a solid
foundation for their AI initiatives. 41% of leaders agree that infrastructure is the area
they need most help with and 39% need help with strategic planning and consulting,
citing specialized components like infrastructure or security as well as broader design
and implementation. Additionally, 43% are predominantly proactive in developing their
AI infrastructure strategy, compared to 16% who are mostly reactive. There’s a clear
opportunity for partners to provide the consultation and expertise companies need to
optimize their infrastructure for AI.

Areas organizations need the most help with

41% 41% 40% 39%

Training and Infrastructure design Security and Strategic planning
support​ and implementation compliance​ and consulting​

Most leaders aim to be proactive

43% 42%
Strategic planning and consulting is desired
by many organizations across all industries.
16% Business leaders that were earlier in their
implementation were more likely to need help
with 42% of AI beginner organizations stating
they needed help. As organizations begin their
journey, they can leverage the consultation of
Proactive A mix of both Reactive partners.
proactive and
reactive

Base: Total (n=900)

Source: A commissioned study conducted by Ipsos on behalf of Microsoft, October 2023

15
Industries bring their own nuances
Manufacturing tends to be the most proactive (51%) in planning for AI infrastructure,
significantly more so than those in healthcare and retail. In such a process-driven
industry, even minor gains in efficiency can bring critical advantages over the
competition. AI technologies, even in its current form, brings giant leaps forward in
operational efficiency.

The entire industry stands to gain significant benefits through process optimizations,
advanced automation, and predictive maintenance. Business leaders across other
industries can follow manufacturing’s example and move quickly to stand up the
right infrastructure for their AI needs. Like any other business strategy decision, these
differences highlight the cross-industry nuances that need to be considered when taking
on any new technology.

Manufacturing organizations aim to be the most proactive

Finance 43%
Healthcare 38%
Retail 39%
Manufacturing 51%
Independent 43%
Software Vendors

Base: Total (n=900)

Source: A commissioned study conducted by Ipsos on behalf of Microsoft, October 2023

16
Start with the trifecta:
performance, security,
and cost
There is no doubt there’s a lot to consider when finding the right AI infrastructure a
business needs, and the speed of changes being brought to the market add further
complexities. As businesses start analyzing vendors and setups, they can look to these
top considerations as a starting-off point: performance, security, and cost.

Top priorities for AI infrastructure

44% 42%
44% 42%
37% 37%

Performance Security and Cost Integration with

and scalability privacy effectiveness existing systems

Base: Total (n=900)

Source: A commissioned study conducted by Ipsos on behalf of Microsoft, October 2023

17
Performance and scalability
With 44% of leaders prioritizing performance and
scalability, particularly in industries managing high-volume
and complex AI workloads like retail, manufacturing,
49%
and independent software vendors, the imperative for AI 44% 52%
infrastructure lies in providing swift and reliable computing
resources to optimize resource utilization, reduce latency,
and scale up and out as needed. Focusing on performance
Independent
and scalability means looking beyond the cost to consider
software
all the benefits that come with AI and maximizing the full vendors
impact of AI infrastructure implementations.

Security and privacy

Security and privacy also take precedence, with 42% of business leaders citing it as a
top priority. This is especially true for finance and healthcare where they must secure
confidential data against unauthorized access, cyber threats, and data breaches while
complying with stringent regulations. While security fell slightly in importance for retail,
manufacturing, and ISV, these industries still rated it as a key priority to address overall.
These factors are essential for ensuring the reliability, efficiency, and effectiveness of AI
infrastructure solutions, and addressing the key challenges that leaders face.

“Security and privacy” tops the list for finance and healthcare

49%
49% 44%
44% 52%
52%

Finance Healthcare

Base: Total (n=900)

Source: A commissioned study conducted by Ipsos on behalf of Microsoft, October 2023

18
 Cost effectiveness
Not surprisingly, cost effectiveness is a priority, with 37% of business leaders across
industries citing getting the value they need from their AI infrastructure and meeting
ROI goals as important. Cost is even more relevant for retail industries.

AI infrastructure Organizational Barriers and blockers

priority challenges for AI adoption

37% 32% 20%

Cost effectiveness ROI is unclear Cost and ROI

(rank 3 of 10) (rank 2 of 13) (rank 4 of 20)

Beyond these top three priorities,

Cost effectiveness is more important to
businesses expect the following leaders in the retail industry (42%) than
from their AI infrastructure: other industries and tops their list of AI
infrastructure priorities.

Other priorities

37% 31% 29% 23% 23% 19%

User-friendly Flexibility Robust data Continuous Support and

tools and and management performance maintenance
interfaces customization monitoring and
optimization

Base: Total (n=900)

Source: A commissioned study conducted by Ipsos on behalf of Microsoft, October 2023

19
 One size does not fit all
Priorities are fluid, shifting to match an organization’s constantly evolving context.
Factors like industry, market, AI maturity level, and platforms create a constantly
evolving environment to navigate. Leaders need to be able to plan for these shifts
in priorities by understanding the organization’s changing context and impact on
implementations.

Level of AI readiness
Leading-edge organizations face different challenges compared to businesses that
are earlier in their AI infrastructure implementations. As an organization moves further
along, flexibility, data management, maintenance, and support compete with earlier
priorities like performance, security, cost effectiveness, and integration.

As new technologies and processes become their standard mode of operation,

priorities change, and it’s important for leaders to proactively plan for these changes.
Understanding that more things become important as your organization progresses will
enable you to pivot priorities quickly to meet changing and expanding needs.
Leading-edge 34%

Priorities that intensify byAIreadiness

beginner level 17%

Robust data management Performance and scalability

Leading-edge 34% Leading-edge Leading-edge 38% 37%

AI beginner 17% Developing AI beginner 51% 47%

Security and privacy

4% Leading-edge Leading-edge38% 37%

Developing AI beginner51%
7% 47%

Base: Total (n=900)

8% Source: A commissioned study conducted by Ipsos on behalf of Microsoft, October 2023

1% 20
Workload type
The AI infrastructure capabilities needed vary based on business model and workloads.
One company may need fully customizable software, services, and computing
while off-the-shelf AI models, services and platforms may be fine for another. As
AI continues to evolve, the needs of companies and even the solutions are rapidly
changing. Because of this, providers are ramping up their offerings to reach a wider
range of needs and providing multiple points at which a customer can come into their
system.

We’ve identified three different categories of customers for AI infrastructure.

AI leaders
‘AI leaders’ have a defined AI strategy and want to lead their market by building their
own innovative, homegrown AI models and applications. They require highly performant
supercomputing infrastructure that can flexibly meet complex storage, network, computing,
and security needs. Their AI workloads are incredibly complex, involve massive models, and
require control at every layer of their AI infrastructure.
Business drivers: Developing an end-to-end AI service, solution or platform, from the ground
up. Require unlimited scalability and the ability to deliver customized user experiences.

AI power users
‘AI power users’ also have a defined AI strategy and are heavily customizing pre-built AI
models, infusing company-specific content and data, and retraining. They need control over
each layer of their AI infrastructure but typically don’t require massive compute power when
working with pre-trained models.
Business drivers: Looking to maximize efficiency and minimize time to market. Save time by
using pre-built models and optimizing for their needs.

AI ready
‘AI ready’ companies want infrastructure that is ready to go so they can focus on defining their
AI strategy. They don’t want to worry about the ins and outs of infrastructure, they are looking
for a scalable, out-of-the-box solution that can support discrete processes now and support
AI growth.
Business drivers: Taking their first steps with AI. Need off-the-shelf solutions.

21
Platform considerations
There is no one-size-fits-all to determining whether a company should be on-premises,
hybrid, or on the cloud – every type of solution has their positives and negatives. With
a myriad of factors at play, ultimately the decision about what is best lies with the
company and their unique situation. For example, on-premises AI infrastructure may
offer more control but require more upfront investments and can be expensive to
maintain, difficult to scale, and hard to keep up to date with the latest technologies.

AI infrastructure in the cloud offers fast deployment, scalability and flexibility, and
typically the best technology available, but some have concerns with security, privacy,
and meeting compliance requirements. Hybrid setups offer the benefits (and drawbacks)
of both but with higher degrees of complexity. Startups may benefit even more from
a cloud set-up due to their small employee size and their naturally increased focus on
getting their product to market as fast as possible. Nonetheless, key themes that came
across for each of these were security and cost effectiveness.

Top three priorities by solution setup

On-premises Hybrid Cloud

• Data security • Security and compliance • Data privacy and security

• Cost effectiveness • Cost effectiveness • Flexibility and scalability

• Existing IT infrastructure • Scalability and elasticity • Cost optimization

integration

For more advanced models, we need GPUs that will be more useful. But for
now, we have some, light-weight models that can be run in CPUs.

The infrastructure requirements will be different depending on the planned

use-case. For example, an early pilot project and actual full-scale
implementation have completely different infrastructure. requirements.

Base: (n=641)
Source: A commissioned study conducted by Forrester Consulting on behalf of Microsoft and NVIDIA, May 2023

22
Half of business leaders said they have a hybrid setup with plans to migrate fully to
the cloud for better IT governance and security, increased productivity and scalability,
more successful deployments, greater innovation, and ROI. For those that are “AI
ready,” a cloud provider may provide a more comprehensive out of the box solution to
get started.

Further than deciding whether to have an on-premises, hybrid, or cloud setup, the
vendor choice is fraught with a range of choices. Key features like having high quality
AI algorithms, ability to manage networking capabilities, integration with open-
source tools, API accessibility, scalability, clear documentation, and multi-cloud/hybrid
enablement came through as the highest requirements.

How would you describe your organization’s current AI workload implementation?

49%
31%
17% 3%

Mix of on-premises Mostly to fully Mix of on-premises On-premises

and public cloud with public cloud and public cloud with
plans to migrate with no plans no plans to migrate
to migrate

What are the main benefits of hosting AI workloads in the cloud?

Better IT governance
Increased productivity
Higher percentage of AI concepts successfully deployed in production
Increased scalability
Increased data security for ML models and data sets

Base: (n=641)
Source: A commissioned study conducted by Forrester Consulting on behalf of Microsoft and NVIDIA, May 2023

23
Industry context
Different industries have different priority nuances. More regulated industries like
finance and healthcare put a higher focus on security and privacy, while manufacturing
and ISVs require strong performance and scalability. Additional factors like solution set-
up and level of AI maturity will also make a difference in both their organizational and
technical priorities.

Top 3 AI infrastructure priorities by industry

Finance Healthcare Retail

1. Security and privacy 1. Security and privacy 1. Performance and

scalability
2. Performance and 2. Performance and
scalability scalability 2. Cost effectiveness
3. Integration with existing 3. Integration with existing 3. Integration with
systems systems existing systems

Independent
Manufacturing Software Vendor
1. P
 erformance and 1. Performance and
scalability scalability

2. Integration with 2. Security and privacy

existing systems
3. Cost effectiveness
3. Security and privacy

Security is a major concern, with so much data available and protections

needed. (Infrastructure let us) use machine algorithms to detect threats in
real time and enhance cybersecurity.

AI lets us shorten the time required for so many things. And it helps our
performance and overall activity.

Base: Total (n=900)

Source: A commissioned study conducted by Ipsos on behalf of Microsoft, October 2023

24
Harnessing the power
of AI now
To help businesses move forward in their AI journey, we recommend four actions to
help navigate the challenges and speed AI production and integration.

Prioritize your AI infrastructure

Infrastructure is at the core of AI innovation. It can determine how fast, how good, how
easy, how groundbreaking, and how engaging an AI application, solution, or platform
will be. Companies should carefully examine their AI goals and strategy and determine
what infrastructure capabilities and platform (on-premises, cloud, hybrid) best fit their
needs today and in the future. It is rare that existing infrastructure can power the
demands and complexity of AI.

Most businesses will need to make changes, either overhauling their existing
infrastructure, choosing a solution provider offering a full-stack AI platform, or
something in between. A company’s AI infrastructure strategy can shape the future of
their business, either accelerating their AI journey or blocking their innovation.

Overcome the skills gap

To overcome the AI skills gap, business leaders need to invest in training and upskilling
their current employees and/or consider bringing in outside talent. Partnering with
an experienced AI solution provider can also help fill the void and deliver employee
training, strategy planning, and AI infrastructure, production, and implementation
support.

25
Make it secure
Security, privacy, and compliance should be at the forefront of any AI and infrastructure
plans. Secure AI is the act of securely designing, developing, and deploying AI and
GenAI capabilities and systems. Follow these best practices:

• Keep user data private and secure.

• Ensure transparency in procedures and emphasize the significance of clearly

communicating sources and criteria for decision-making.

• Ensure security is built-in from inception to deployment of the AI system’s

lifecycle.

• Keep risk at the forefront when designing interfaces and processes.

Find a partner
Across all industries, business leaders expressed a need for help with strategic planning
and consulting as well as training and support from AI solution providers. Leading-
edge organizations partner with AI experts to help plan, build, and integrate AI into
their business. Companies of any size and at any stage can benefit from a strategic AI
solution provider. Forming a partnership with a proven AI solution provider can be key
to accelerating AI production and staying competitive.

26
Take the next steps on your
AI transformation journey
Explore how Microsoft Azure is redefining cloud infrastructure to prepare every
business for AI by providing the world-class technology for AI workloads and doing
so sustainably and responsibly.

Get strategic guidance and insights on AI innovation, tailored for business leaders.

Learn how businesses are balancing performance, efficiency, and cost with Azure AI
infrastructure.

© 2024 Microsoft Corporation. All rights reserved. This document is provided "as-is". Information and views
expressed in this document, including URL and other internet Web site references, may change without
notice. You bear the risk of using it. THis document does not provide you with any legal rights to any
intellectual property in any Microsoft product.

27
Research methodology
In May 2023, we commissioned Forrester Consulting to evaluate the current state of AI
among IT directors and decision makers in North America, Europe, and Asia Pacific.

We further explored this topic through a second study with Ipsos in September 2023
among technical and business leaders, developers, and data professionals who are
early majority adopters of technology across 3 markets (US, Germany, and India) and
5 different industries (finance, healthcare, retail, manufacturing, and independent
software companies).

Forrester Consulting Research Ipsos Research

Fielding: May – June 2023 Fielding: September - October 2023

Participants: n=641 Director and above in IT Participants: n=900 ITDMs who are in the
with responsibilities in AI workloads, cloud early majority of technology adopters at
infrastructure. their organizations.

Company Size: 1000+ FTE in North America, Company Size: ISVs = 50+ employees
500+ in all other countries All other industries = 500+ employees

Countries: North America (US, Canada), Countries: US (n=500), Germany (n=200),

EMEA (UK, Germany, France), APAC India (n=200)
(Australia, New Zealand)
Industries: Finance (US n=100, Germany/
Industries: All industries, including India n=40); Healthcare (US n=100,
automotive, manufacturing, oil and gas, Germany/India n=40); Retail (US n=100,
financial services, public sector, universities, Germany/India n=40); Manufacturing (US
bio life sciences. n=100, Germany/India n=40); ISV (US n=100,
Germany/India n=40)
ISVs are independent software vendors
or software houses that develop software for
broad commercial distribution, including SaaS.

28