Building a Foundation
for AI Success:
A Leader’s Guide
September Building a Foundation for AI Success: ii
2023 A Leader’s Guide
Table of
Contents
01 / 15 /
Introduction Getting Started
02 / 18 /
Definitions Conclusion
03 / 19 /
Five Pillars of AI Success Sources
06 Business Strategy
07 Technology Strategy
08 AI Strategy and Experience
10 Organization and Culture
13 AI Governance
September Building a Foundation for AI Success: 1
2023 A Leader’s Guide
Introduction In the nearly 70 years since John McCarthy
coined the term “artificial intelligence,”
we’ve seen wave after wave of technological
innovation, from the rise of personal
computing to the internet, mobile devices,
the cloud, and now, generative AI.
Each of these shifts has created new
opportunities and new questions for leaders.
How can these technologies help my
organization thrive? What new possibilities
could they offer? How should I organize
for optimal impact? And, perhaps most
importantly, how can I use these technologies
in a way that promotes trust among customers,
citizens, patients, partners, shareholders, and
the public?
In this paper, we’ll share insights—from
customers, partners, analysts, AI leaders inside
and outside of the company, and from our own
experience—about the five categories that,
collectively, form a foundation for building
sustainable value with AI. They are:
• Business strategy
• Technology strategy
• AI strategy and experience
• Organization and culture
• AI governance
Finally, we’ll offer suggestions for pragmatic
next steps to accelerate your progress, whether
you’re just getting started, developing your
organizational expertise, or well along the path.
September Building a Foundation for AI Success: 2
2023 A Leader’s Guide
Definitions Artificial Intelligence (AI) (1950s): the theory
and development of computer systems
that are able to perform tasks that normally
require human intelligence, such as visual
perception, speech recognition, decision-
making, and translation between languages1
Machine Learning (1990s): a subset of AI and
computer science where algorithmic models
are trained to learn from existing data to
make decisions or predictions
Deep Learning (2010s): a machine learning
technique that uses layers of neural networks
to process data and make decisions
Generative AI (2020s): a type of AI
technology that uses algorithmic models
to create new written, visual, and auditory
content when given prompts or existing data
September Building a Foundation for AI Success: 3
2023 A Leader’s Guide
Five
Pillars
of AI
Success
September Building a Foundation for AI Success: 4
2023 A Leader’s Guide
While the term “artificial intelligence” natural ways,” says Kevin Scott, Executive
has been around since the 1950s, the Vice President and Chief Technology Officer
pace of AI innovation has accelerated in at Microsoft.
the past several years, thanks to plentiful
data, greater access to computing Given the pace of AI innovation, the most
power, and the size and sophistication frequent questions we hear from customers
of algorithmic models.2 are: Where do I go from here? How do I
create the most impact? What does success
AI is fundamentally different from look like?
technologies that have come before it,
both in its approach and its capabilities. It While there is no one answer for all
is based on probabilities, rather than rules, organizations, we’re starting to see best
and “learns” from data without requiring practices emerge across five discrete
explicit instructions. AI makes it possible categories, which we refer to here as the
for computers to perform certain tasks five pillars of AI success. (See Figure 1.)
that previously only humans could do, such
as visual perception, speech recognition,
decision-making, and language translation.3
Generative AI takes that one step further. It
can be used for, among other things, content
creation, summarization, and question
answering, as well as the creation of images
based on text prompts, all using everyday
language. “For the first time, there will be
a way for every human being to express
their intent to a computing device in very
September Building a Foundation for AI Success: 5
2023 A Leader’s Guide
Figure 1: Five Pillars of AI Success
Stages
Business strategy Stage 1:
Clearly defined and prioritized business Exploring
objectives, use cases, and measurement Learning about AI and
experimenting with it
of AI value
in some parts of
the organization
Technology strategy
An AI-ready application and data
platform architecture, aligned parameters Stage 2:
for build vs. buy decisions, and plans for Planning
Actively assessing,
where to host data and applications to
defining, and planning
optimize outcomes AI strategy across
the organization
AI strategy and experience
Stage 3:
A systematic, customer-centric approach
Formalizing
to AI that includes applying the right
Formalizing, socializing,
model to the right use case and and executing on AI strategy
experience in building, testing, and across the organization and
in multiple business
realizing AI value across multiple business
units while starting
units, use cases, and dimensions to realize value
Organization and culture Stage 4:
A clear operating model, leadership Scaling
support, change-management process, Delivering both
incremental and
access to continuous learning and
new value across
development, and strong relationships the organization
with diverse subject-matter experts
AI governance Stage 5:
Implementation of processes, controls,
Realizing
Realizing consistent
and accountability structures to govern AI value across the
data privacy, security, and responsible organization and in
use of AI multiple business units
September Building a Foundation for AI Success: 6
2023 A Leader’s Guide
Business Strategy
Clearly defined and prioritized business objectives,
use cases, and measurement of AI value
One of the biggest drivers of AI success is Nineteen percent of respondents cited
the degree to which the organization has “unable/hard to measure the value,” and
defined and prioritized business objectives, 19 percent cited “lack of understanding AI
use cases, and how it will measure value. benefits and uses.” 4
This is particularly important given the
wide applicability of AI to so many different In Quick Answer: What Is the True Return on
needs, such as process optimization, content AI Investment? Gartner stated: “Enterprises
generation, summarization, procurement, do not achieve maximum leverage from
supply-chain optimization, and more. Success artificial intelligence investments, despite
requires rigorous focus on strategic goals increased spending. Executive leaders must
as well as a growth mindset to embrace become keen and discerning creators of
challenges and learn from failure. AI investment strategies in order to obtain
optimum value from AI initiatives,” and that
“Rather than starting by asking what AI can “the best return yield from AI investment
do, we need to turn the telescope around will come from an extensive portfolio of
and ask, ‘What are you trying to do in your AI, guided by an expansive and evolving
business, and how can AI help?’” says Jason investment thesis that is aligned to strategic
Price, Director of Specialist Management priorities and helps to allocate resources
at Microsoft. based on business impact. Organizations
that follow a portfolio management plan to
In fact, the Gartner® 2022 AI Use-Case determine most AI use cases are 2.4 times
ROI Survey states that the “main barriers more likely to reach ‘mature’ levels of AI
preventing implementation of AI are implementation.”5
unable/hard to measure value and lack
of understanding AI benefits and uses.”
September Building a Foundation for AI Success: 7
2023 A Leader’s Guide
Technology Strategy
An AI-ready application and data platform architecture, aligned parameters
for build vs. buy decisions, and plans for where to host data and applications
to optimize outcomes
The pace of AI innovation has captured the As you work to resolve these questions,
imaginations of people around the world. the first step is to choose an application
It has also intensified many of the biggest and data platform architecture that will
questions leaders face when seeking to meet your organization’s requirements.
optimize AI value, such as: Your architecture will determine the
technologies you need, whether you buy a
• Do I have the infrastructure required for prebuilt solution, build it yourself, or opt for
AI applications to access data securely, a combination.
quickly, and at scale?
“You can’t democratize AI if you don’t have
• Based on my top-priority use cases,
an architecture that connects everyone
should I buy, build, or modernize AI
across the company,” says Andy Markus,
applications?
Chief Data Officer at AT&T. “The cloud makes
• How should I determine whether to host that possible.”
data and AI applications on premises or
in the cloud?
September Building a Foundation for AI Success: 8
2023 A Leader’s Guide
AI Strategy and Experience
A systematic, customer-centric approach to AI that includes applying the right model to
the right use case and experience in building, testing, and realizing AI value across multiple
business units, use cases, and dimensions
Customer-centricity, and taking a systematic
approach to AI, are both emerging as
77 %
key contributors to AI success. The 2023
Gartner® report Survey Analysis: AI-First
Strategy Leads to Increasing Returns found
that “41 percent of mature AI organizations
use customer success-related business of mature
metrics,” while “the strategic importance organizations adopt
of AI techniques has once again been
an AI-first strategy,
confirmed by our respondents. 77 percent
of mature organizations adopt an AI-first
systematically
strategy, systematically considering AI for considering AI for
every use case.”4 every use case.4
Another critical driver of AI success is
applying the right model to your use case—
in other words, using the right tool for the
right job—to solve specific problems and
realize value. This applies whether you’re
buying applications or building or your own.
September Building a Foundation for AI Success: 9
2023 A Leader’s Guide
Additional factors that tend to correlate with
an organization’s AI success include:
• The number of AI use cases deployed.
• The length of time they’ve been in use.
• The degree to which they have scaled
across the business.
• The degree of value they have
generated, measured by productivity,
revenue, or another metric.
These factors can also reveal potential
barriers to success. One of the most
common examples is the “perpetual proof of
concept” loop, which tends to point to gaps
in alignment between projects and valued
business outcomes. Finally, the degree to
which AI is democratized throughout the
organization—extending AI capabilities and
tools from a small group of experts to the
organization as a whole—is also a leading
indicator of success.
September Building a Foundation for AI Success: 10
2023 A Leader’s Guide
Organization and Culture
A clear operating model, leadership support, change-management process, access
to continuous learning and development, and strong relationships with diverse
subject-matter experts
In conversations with customers and Salim Naim, Director of Specialist
partners, organization and culture frequently Management at Microsoft, says one of
emerge as critical factors for success. the key questions to consider “is whether
your operating model is geared towards
Operating model just experimentation and centralized, or
According to Gartner: “The pace of AI whether it is designed to be embedded in
technology maturation and diverse every aspect of the business.” He advises
approaches make it difficult to capture and organizations to ensure they’re taking an
sustain value from AI initiatives. Effective inclusive approach to developing their
AI operating models that leverage current operating model. Leaders should ask
investments in people, processes, and themselves: “What should my operating
technologies enable IT leaders to drive model be that allows business units
successful AI initiatives.”6 It can mean the and different geographies in global
difference between AI projects that are organizations to adopt it?”
viewed as science experiments and those
that become significant value-drivers. Leadership support
In discussions with customers and
partners, and in our own experience of
AI transformation, we have found that
organizations that derive the most value
from AI are typically those whose leadership
recognizes and supports the opportunities
of AI with words, resources, and actions.
September Building a Foundation for AI Success: 11
2023 A Leader’s Guide
This may start as an initiative from the top
or as a grassroots effort that gradually
gains momentum. Either way, says Andreas
Nauerz, Chief Technical Officer at Bosch
Digital, “You need an atmosphere which
encourages people to experiment and
learn, even from failure. This is especially
crucial considering the pace of technological
advancement.”
To learn more about how Microsoft
customers are approaching AI
transformation in their organization and
culture, read “Lessons from Enterprise AI
Pioneers” in the Wall Street Journal.7
Change management
An organization’s ability to manage change
is also a critical driver of AI success. “You very
quickly learn that by the time you succeed
with something, it’s already outdated,” says
Mikkel Bernt Buchvardt, Director of Data and
Analytics at SEGES Innovation. He suggests
embracing this reality, rather than letting it
slow you down. “You can keep gold-plating
your methods, or you can make it good
enough to deliver some value.”
September Building a Foundation for AI Success: 12
2023 A Leader’s Guide
Skill-building and learning across a spectrum of competencies to
Access to skill development, continuous ensure that AI projects truly serve business
learning, and certifications are also key. objectives. As Andy Markus says: “The very
But according to Salim Naim, AI success first step of the journey is not even technical.
isn’t simply skill acquisition—it needs to It’s to establish a great partnership with the
become a more sustainable capability within business. The number one goal is to deliver
your organization. “As you mature, you go value to the company and to our customers.
beyond what you solve to how well you Sure, we’re technologists, and we can get
solve it,” he says. really jazzed about doing cutting-edge
things with technology. But the ultimate
Strong relationships with reason we’re here is to deliver value for our
subject-matter experts company and to our customers.”
Access to technologists with the right skills
in the right roles is also fundamental to For more insights on AI and the world of
success. But it’s just as important to foster work, see the Microsoft WorkLab.
relationships with subject-matter experts
“The very first step of the journey is not
even technical. It’s to establish a great
partnership with the business. The
number one goal is to deliver value to
the company and to our customers.”
Andy Markus
AT&T Chief Data Officer
September Building a Foundation for AI Success: 13
2023 A Leader’s Guide
AI Governance
Implementation of processes, controls, and accountability structures
to govern data privacy, security, and responsible use of AI
“Don’t ask what AI customer commitments.10 “Every
computers can do. Ask organization that creates or uses AI systems
will need to develop and implement its own
what they should do.”
governance systems,” he said.
Microsoft President Brad Smith wrote those
Organizations seeking to reap the greatest
words about the ethics of AI in a book he
benefit from AI must develop their
co-authored in 2019. “This may be one of the
understanding of the data governance,
defining questions of our generation,” the
security, and responsible AI implications
authors declared.8
of their decisions, with regard to both risks
Four years later, AI—and its relationship to and opportunities.
trust, data privacy, and security—is on the
Since 2017, Microsoft has been sharing
minds of people and organizations around
expertise, providing training curriculum,
the world. “Consumer faith in cybersecurity,
and creating dedicated resources to support
data privacy, and responsible AI hinges
responsible use. “A theme that is core to our
on what companies do today,” a recent
responsible AI program and its evolution
McKinsey report stated.9
over time is the need to remain humble and
As with any consequential new technology, learn constantly,” says Natasha Crampton,
AI must be built on a foundation of security, Chief Responsible AI Officer at Microsoft.
risk management, and trust. “Ensuring “Responsible AI is a journey, and it’s one
the right guardrails for the responsible that the entire company is on.”11
use of AI will not be limited to technology
For insight on Microsoft’s approach to
companies and governments,” wrote
Responsible AI, read “The building blocks of
Antony Cook, Corporate Vice President and
Microsoft’s responsible AI program.”12
Deputy General Counsel at Microsoft, in a
recent blog post announcing Microsoft’s
September Building a Foundation for AI Success: 14
2023 A Leader’s Guide
A theme that
is core to our
responsible AI
program and
its evolution
over time is the
need to remain
humble and
learn constantly.
Responsible AI is
a journey, and it’s
one that the entire
company is on. 11
Natasha Crampton
Chief Responsible AI Officer, Microsoft
September Building a Foundation for AI Success: 15
2023 A Leader’s Guide
Getting
Started
September Building a Foundation for AI Success: 16
2023 A Leader’s Guide
Getting Started
Use this guide to help identify priority areas as
you define your AI strategy and roadmap.
Suggested next steps
• Define and prioritize business objectives such as customer experience,
1 productivity, revenue growth, employee experience, and other key goals.
Business • Determine how you will measure the value of those objectives.
strategy
• Identify and prioritize AI use cases that support your goals.
• Build a portfolio management plan to help guide your investments.
• Based on your top-priority use cases, determine whether to buy, modernize, or
2 build applications.
Technology • Assess whether you have the infrastructure for AI applications to access data
strategy securely, quickly, and at scale.
• Consider the scalability and performance implications of hosting data and AI
applications on premises or in the cloud.
• Ensure your cloud infrastructure is built to run large AI workloads and deliver
reliability at scale.
• Evaluate your organization’s Zero Trust security posture.
• Explore how to use AI for improving security, in terms of deploying and protecting
organizational assets, developing and maintaining policies and procedures, and
monitoring and responding to incidents or emerging threats.
September Building a Foundation for AI Success: 17
2023 A Leader’s Guide
Suggested next steps
• Familiarize yourself with generative AI use cases and how they might support your
3 business needs.
AI strategy and • Develop a systematic process to consider AI for every use case.
experience
• Assess the number of business units and processes, length of time in production,
and age of deployments in your organization to reveal patterns that may point to
opportunities or blockers.
• Build intelligent apps on your data to improve the intelligence and relevance of
model outputs.
• Consider using Microsoft 365 Copilot, or build your own copilot to accelerate
learning and time to value.
• Define your operating model for AI.
4
• Secure—or develop a plan to secure—leadership support backed by resources.
Organization
and culture • Develop strong relationships with a diverse range of subject-matter experts in the
business.
• Strengthen your organization’s ability to manage change.
• Identify and implement the right learning and skill-building paths in place.
• Approach AI as a sustainable capability within your organization and culture.
• Review and share resources on responsible use of AI to identify the models and
5 approaches that best suit your organization.
AI governance • Consider the enablement model that best fits your needs, such as hub-and-spoke,
centralized, or distributed.
• Consider the principles of secure AI and how to ensure your data is protected end
to end from platform to applications and users.
• Consider the processes, controls, and accountability mechanisms that may be
required to govern the use of AI, and how AI may affect data privacy and
security policies.
September Building a Foundation for AI Success: 18
2023 A Leader’s Guide
Conclusion Given the speed of
AI innovation and
the uniqueness of
every organization,
no single framework
could possibly
account for all
organizations,
regions, or
circumstances. But
we hope that these
pages have helped
you identify potential
areas of strength
and opportunities
to accelerate your AI
innovation journey.
September Building a Foundation for AI Success: 19
2023 A Leader’s Guide
Sources
1
Oxford Reference. Overview: Artificial Intelligence. 7
Wall Street Journal. “How Leaders Plan for AI Success.”
Retrieved August 14, 2023, from Artificial intelligence - Paid Program: Lessons From Enterprise AI Pioneers
Oxford Reference. (wsj.com).
2
Kelly, Kevin (2016). The Inevitable: Understanding the 12 8
Smith, Brad, Browne, Carol Ann, and Gates, Bill
Technological Forces That Will Shape Our Future. Viking. (Foreword) (2019). Tools and Weapons: The Promise and
Peril of the Digital Age. Penguin.
3
Oxford Dictionaries. Retrieved August 25, 2023, from
artificial intelligence - definition of artificial intelligence in 9
Excerpted from “Why digital trust truly matters,”
English from the Oxford dictionary (ucl.ac.uk). September 2022, McKinsey & Company,
www.mckinsey.com. Copyright (c) 2023 McKinsey &
4
Gartner. AI-First Strategy Leads to Increasing Returns. Company. All rights reserved. Reprinted by permission.
Retrieved August 14, 2023, from Survey Analysis: An
AI-First Strategy Leads to Increasing Returns, by Erick Microsoft. Retrieved August 16 2023 from Announcing
10
Brethenoux, Published June 21, 2023 . GARTNER is a Microsoft’s AI Customer Commitments - The Official
registered trademark and service mark of Gartner, Inc. Microsoft Blog.
and/or its affiliates in the U.S. and internationally and is
used herein with permission. All rights reserved.
11
Microsoft. Retrieved August 16, 2023 from Reflecting
on our responsible AI program: Three critical elements
5
Gartner, Quick Answer: What is the True Return on AI for progress - Microsoft On the Issues.
Investment? By Ethan Cohen, Afraz Jaffri, Published
April 26, 2023. (gartner.com). Survey_Analysis_ Microsoft. Retrieved September 8, 2023 from The
12
An__795644_ndx.pdf. building blocks of Microsoft’s responsible AI program -
Microsoft On the Issues.
6
Gartner, Quick Answer: How Should CXOs Structure AI
Operating Models? By Chirag Dekate, Soyeb Barot, Sumit
Agarwal, Anthony Mullen, Published February 27, 2023.
©2023 Microsoft Corporation. All rights reserved. This document is provided “as-is.”
Information and views expressed in this document, including URL and other Internet
website 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. You may copy and use this document for your internal,
reference purposes.
