2024_Sharholder_Letter
2024_Sharholder_Letter
2024_Sharholder_Letter
Fiscal year 2024 was a pivotal year for Microsoft. We entered our 50th year as a company
and the second year of the AI platform shift. With these milestones, I’ve found myself
reflecting on how Microsoft has remained a consequential company decade after decade in
an industry with no franchise value. And I realize that it’s because—time and time again,
when tech paradigms have shifted—we have seized the opportunity to reinvent ourselves to
stay relevant to our customers, our partners, and our employees. And that’s what we are
doing again today.
Microsoft has been a platform and tools company from the start. We were founded in 1975
with a belief in creating technology that would enable others to create their own. And, nearly
50 years later, this belief remains at the heart of our mission to empower every person
and every organization on the planet to achieve more.
This year, we moved from talking about AI to helping our customers translate it into real
outcomes—one person, one organization, one institution, and one country at a time. We
have made remarkable progress on this front across every industry. For example:
Coles is generating 1.6 billion daily AI predictions across 850 Australian stores,
ensuring every shopper finds what they need.
Unilever is performing thousands of simulations with AI in the time it would take to
run tens of laboratory experiments, as it accelerates its product development.
Developers at Itaú, Brazil’s largest private bank, are coding more efficiently using our
AI pair programmer, GitHub Copilot.
Khan Academy is making tutoring more accessible for students and helping teachers
plan more creative lessons, using our small language model Phi.
Aquafarmers in Indonesia are improving their yields, thanks to an app built with the
Azure OpenAI Service, as well as Azure IoT.
In Kenya, street vendors now have access to credit for the first time, thanks to M-
Kopa, a social enterprise using Azure ML to do its forecasting.
And enterprise customers and their employees around the world, from Amgen and
Disney, to Finastra and Vodafone, are using Microsoft 365 Copilot to become more
creative and productive.
Financially, the year was also marked by record performance. We delivered over $245 billion
in annual revenue, up 16 percent year-over-year, and over $109 billion in operating income,
up 24 percent.
Going forward, we are focused on three priorities: First, prioritizing fundamentals, with
security above all else. We launched the Secure Future Initiative (SFI) this year, bringing
together every part of our organization to advance cybersecurity protection. Second, driving
trustworthy AI innovation across our entire portfolio while continuing to scale our cloud
business. And, finally, managing our cost structure dynamically to generate durable, long-
term operating leverage. All three priorities are critical to our ability to continue thriving as a
company as we raise the bar on our operational excellence, with a focus on continuous
improvement across everything we do.
AGE OF AI
If we go back 70 years to the beginning of modern computing, our industry has had two
dreams: First, can computers understand us instead of us having to understand computers?
And second, as we digitize more of the world—including people, places, and things—can
computers help us reason, plan, and act more effectively using all that information? Over the
past year, we have had breakthroughs on both fronts.
The core underlying force behind these breakthroughs is scaling laws. Just like Moore’s Law
drove the information revolution, the scaling laws of deep neural networks (DNN) and
transformers are driving today’s AI revolution. Up until the DNN inflection point, progress in
compute was keeping up with Moore’s Law—doubling every two years. But we have now
started to see progress in AI performance double roughly every six months.
There are three capabilities coming together because of these scaling laws. First, we have a
new natural user interface that is multimodal. It supports speech, images, and videos—both
as input and output. We have memory that retains important context, recalling both our
personal knowledge and data across devices, apps, and the web. And, finally, we have new
reasoning and planning capabilities that help us understand complex context, complete end-
to-end tasks on our behalf, and reduce our cognitive load.
This new world is being defined by a rich tapestry of AI agents, which can take action on our
behalf, including personal agents across work and life, business process agents, and cross-
organizational ones. These agents will be able to work in concert as a new input to help
make small businesses more productive, make multinationals more competitive, make the
public sector more efficient, and improve health and education outcomes broadly.
Microsoft has built three leading platforms to help our customers maximize their opportunity
in this emerging agentic era: Copilot, which you can think of as the new UI for AI—the
human interface for this agentic world; the Copilot stack, which brings together
infrastructure, data, and app services to help customers build their own copilots and agents
for their own business processes; and a new category of Copilot devices that are purpose-
built for this new era, including the Copilot+ PCs we introduced this year.
OUR OPPORTUNITY
The innovation we have driven over the past year matters only if we translate it into
enduring value for our customers. That’s why, across our tech stack, we are focused on
helping people and organizations realize the benefits of AI.
Infrastructure
This year, we expanded our cloud and AI capacity, announcing new investments across five
continents. These are long-term assets to drive new growth for the next decade and beyond,
and ensure communities around the world have access to the compute they need to drive
economic growth in this new era.
Our cloud now also offers top performance for AI training and inference and the most diverse
selection of AI accelerators, including the latest from AMD and NVIDIA, as well as our own
first-party silicon, Azure Maia, which we introduced last November.
More broadly, we continued to see sustained revenue growth from migrations as customers
turn to Azure. Azure Arc is helping customers streamline their transition, as they secure,
develop, and operate workloads with Azure services anywhere. We have 36,000 Arc
customers, up 90 percent year-over-year. And we remain the hyperscale cloud of choice for
SAP and Oracle workloads.
Data & AI
AI models are now key building blocks for every application. And with Azure AI, we are
building out the app server for the AI age, providing access to the most diverse selection of
models to meet customers’ unique cost, latency, and design considerations. We offer leading
frontier models, thanks to our strategic partnership with OpenAI. With Phi-3, which we
announced in April, we offer a family of powerful, small language models. And, with Models
as a service, we provide API access to third-party models, including the latest from Cohere,
Meta, and Mistral. In total, we have over 60,000 Azure AI customers, up nearly 60 percent
year-over-year. This year, we also announced a partnership with G42, which will run its AI
applications and services on our cloud, as we collaborate to bring our latest AI technologies
to the United Arab Emirates and other countries.
AI does not get created without data. At the data layer, we are fundamentally rethinking
what it means to be an analytics database or an operational data store in the world of AI.
Our Microsoft Intelligent Data Platform provides customers with the broadest capabilities
spanning databases, analytics, business intelligence, and governance—along with seamless
integration with all our AI services. And Microsoft Fabric, our AI-powered, next-generation
data platform we made generally available this year, now has over 14,000 paid customers
who can go from data, to insights, to action—all within the same unified SaaS solution.
From GitHub to Visual Studio, we have the most comprehensive developer tools. GitHub
Copilot had a breakout year, as it became standard issue for developers in every industry.
We now have more than 1.8 million paid subscribers and over 77,000 enterprise customers,
up 180 percent year-over-year. They are realizing productivity gains of up to 55 percent
while staying in their flow and bringing the joy back to coding. This year, we also introduced
Copilot Workspace, a Copilot-native developer environment, which helps any developer go
from idea, to code, to software—all in natural language.
We are also integrating generative AI across Power Platform, enabling anyone to use natural
language to create apps, automate workflows, or build a website. In total, we now have 48
million monthly active users of Power Platform, up 40 percent year-over-year.
Modern work
Microsoft 365 Copilot is becoming a daily habit for knowledge workers, transforming their
work, workflow, and work artifacts. Adoption has been faster than any other new Microsoft
365 suite. And employees at nearly 60 percent of the Fortune 500 now use Copilot to
complete tasks faster, hold more effective meetings, and automate business workflows and
processes. In fact, internal and external studies show as much as a 70 percent improvement
in productivity using generative AI for specific work tasks. And early Microsoft 365 Copilot
users were 29 percent faster in a series of general tasks like searching, writing, and
summarizing.
And we’re going further, bringing the Web plus Work plus Pages together as the new AI
design system for knowledge work. With Pages, which we just announced last month, you
can take any information from the web or your work and turn it into a multiplayer, AI-
powered canvas. You can ideate with AI and then easily share what you create for
collaboration with other people.
And with Copilot Studio, customers can extend Copilot with agents and build their own
agents that proactively respond to data and events from their own first- and third-party
business data. To date, 50,000 organizations have used it. And, just this week, we
announced new capabilities that will make it possible for customers to build autonomous
agents using Copilot Studio.
Microsoft Teams remains essential to how hundreds of millions of people meet, call, chat,
collaborate, and do business. This year, we rolled out to all customers a new version that is
up to two times faster while using 50 percent less memory. And Teams Premium surpassed 3
million seats, up nearly 400 percent year-over-year, as organizations chose it for advanced
features like end-to-end encryption and real-time translation.
Business applications
We’re using this AI moment to redefine our role in business applications, too. Dynamics 365
once again took share, as organizations use our AI-powered apps to transform their
marketing, sales, service, finance, and supply chain functions.
And we are expanding our total addressable market by integrating Copilot into third-party
systems as well. Our new Dynamics 365 Contact Center infuses generative AI throughout
the contact center workflow in a customer’s existing CRM.
We are also extending Copilot to specific industries, including healthcare. With DAX Copilot,
more than 400 healthcare organizations are increasing physician productivity and reducing
burnout. On average, clinicians save more than five minutes per patient encounter. And 77
percent say it also improves documentation quality.
Security
As I mentioned earlier, security underpins every layer of our tech stack. We are doubling
down on our Secure Future Initiative, as we implement our principles of secure by design,
secure by default, and secure operations. And we are focused on making continuous
progress across the six pillars of the initiative: protect tenants and isolate production
systems; protect identities and secrets; protect networks; protect engineering systems;
monitor and detect threats; and accelerate response and remediation. As part of this
commitment, all Microsoft employees now have security as a “core priority,” holding each
one of us accountable for building secure products and services.
We are continuously applying what we are learning and translating it into security innovation
for our customers. A great example is Copilot for Security, which we made generally
available this year. It brings together LLMs with domain-specific skills informed by our threat
intelligence and 78 trillion daily security signals to provide security teams with actionable
insights.
This year, we introduced an entirely new category of Windows PCs engineered to unleash
the power of distributed AI across the cloud and edge. Copilot+ PCs are the fastest, most AI-
ready Windows PCs ever built. They include a new system architecture designed to deliver
best-in-class performance and breakthrough AI experiences. And we are working across our
entire ecosystem to bring these to life, including with AMD, Intel, and Qualcomm, along with
our OEM partners.
With Copilot, we’re taking the first steps toward creating an AI companion, one that’s always
by your side, helping you feel smarter and more supported through natural conversations.
The refreshed Copilot app we introduced earlier this month delivers a more intuitive design
with more digestible, speedy, and fluent answers. It now adapts to you with a warm tone
and a distinct style, providing not only information but encouragement, feedback, and
advice as you navigate life’s everyday challenges—no matter how big or small. And we’re
adding advanced capabilities like Voice and Vision that make it both more useful and more
natural.
We also continue to apply generative AI to pioneer new approaches to how people search
and browse. Microsoft Bing and Edge both took share again this year. And we introduced
Copilot Pro, providing access to the latest models for quick answers and higher-quality image
creation, and access to Copilot for Microsoft 365 Personal and Family subscribers.
Thousands of news and entertainment publishers trust us to reach new audiences with
Microsoft Start. And we are also helping advertisers increase their ROI. Copilot in Microsoft
Ad Platform helps marketers create campaigns and troubleshoot using natural language.
Gaming
We are bringing great games to more people on more devices. With our acquisition of
Activision Blizzard King, which closed October 2023, we’ve added hundreds of millions of
players to our ecosystem. We now have 20 franchises that have generated over $1 billion in
lifetime revenue—from Candy Crush, Diablo, and Halo, to Warcraft, Elder Scrolls, and Gears
of War. And with Xbox cloud gaming, we continue to innovate to offer players more ways to
experience the games they love—where, when, and how they want. Finally, we brought four
of our fan-favorite titles to Nintendo Switch and Sony PlayStation for the first time, as we
continue to extend our content to new platforms.
OUR MISSION
Although we have made outstanding progress over the past year, we cannot take our
permission to innovate—let alone operate—for granted. It is something we must earn.
We always say Microsoft will do well only if the world around us does well. And that’s why we
are focused on four enduring commitments. They keep us grounded, serving as a guide as
we make decisions, pushing us to ask critical questions to ensure the technology we create
benefits everyone on the planet, as well as the planet itself.
This year, we partnered with 375,000 nonprofits globally, providing $4.7 billion in donated
and discounted technology—to help them build capacity, drive efficiencies, innovate, and
increase their impact as they take on the world’s greatest challenges. Together with our
partners, we also made significant strides in skilling. We have trained and certified 14.1
million people across 202 countries in digital skills as of June 2024. Over 80 percent of the
learners were from groups furthest from opportunity in the digital economy.
And our work doesn’t stop there. We must do more to accelerate the equitable adoption of
AI and narrow the digital divide. That’s why we’ve launched new AI skilling initiatives in
partnership with governments, educational institutions, industry, and civil society—to help
millions of people learn to use generative AI, develop AI tools, and lead AI adoption. This
includes people from underserved communities, with a focus on young people, women, rural
communities, and the Global South, as well as employees of social impact organizations.
Our Secure Future Initiative advances how we design, build, test, and operate our
technology to ensure we deliver solutions that meet the highest possible standard of
security. Our first SFI Progress Report highlights updates spanning culture, governance,
technology, and operations, but we recognize that our work on security is never complete.
We must and will do more. Our promise is to continually learn, improve, and adapt to the
evolving needs of an increasingly complex security landscape.
We are focused on building AI that is trustworthy, meaning that it is secure, safe, and
private. Our responsible AI practices, grounded in our foundational AI principles, help ensure
we do this from the beginning. And we’re building on this commitment by introducing new
product capabilities across our tech stack, ensuring that both our customers and developers
are safeguarded at every level. Ultimately, these commitments and capabilities are key to
fulfilling our mission. The world is looking to us to help defend and protect them, and we
take that responsibility seriously.
In May, we published our first AI Transparency Report, which outlines how we build
generative applications responsibly and support our customers. We also released Goals and
Governance: Goals and Lessons for AI, which draws lessons from other globally governed
technologies like civil aviation and nuclear power. Through our Accelerating Foundation
Models Research program, we’ve made grants to hundreds of projects in AI safety and
alignment research, AI-driven scientific discovery, and beneficial applications of AI. And we
launched our Global Perspectives Responsible AI Fellowship program, designed to center the
voices of AI experts from the Global South and enable us to better understand AI’s impact on
developing countries.
Finally, we continue our work to create safe experiences online and protect customers from
illegal and harmful content and conduct. To bolster our efforts to prevent child sexual
exploitation and abuse risks, we have made new commitments to safety by design in our AI
services, joined the Tech Coalition’s Lantern Program, and proposed concrete actions that US
policymakers can take to protect the public through regulatory and policy measures.
Over the past year, as regulators increasingly required greater transparency and process
consistency across corporate supply chains and human rights efforts, we ensured
compliance with reporting and due diligence directives. Going forward, we’ll continue to
respect global human rights and laws and take steps to mitigate the impact of our
operations and our technology on the people in our value chain.
As we build and deploy more AI solutions, connectivity and accessibility are foundational.
Since 2017, we’ve extended access to affordable high-speed internet to over 100 million
people, including nearly 40 million in Africa. And we remain focused on building inclusive,
accessible AI that empowers people across the spectrum of disability.
More than 4 billion people will vote this year in their respective elections, making it the
biggest election year in history. At the Munich Security Conference in February, we came
together with others across the tech sector and pledged to help prevent deceptive AI
content from interfering with global elections. As part of this pledge, we have worked to
empower campaigns, candidates, election officials, and voters to understand the risks of
deceptive AI in elections and to take steps to protect themselves and democracies. To date,
we’ve conducted deepfake trainings in over 20 countries. And our corresponding public
awareness campaign has reached over 355 million people.
But we also recognize the resource intensity of the infrastructure needed to yield these
benefits, which is why we’re advancing the sustainability of AI, from design to construction
to operations, all while working to improve the efficiency of these technologies. We’re also
investing in innovation through Microsoft’s Climate Innovation Fund and our AI for Good Lab,
advancing research and advocating for policies that can drive global impact.
In our latest Environmental Sustainability Report, we shared progress toward our 2030
commitments. We’re on track in several areas. Where we’re not yet on track, we’re
mobilizing to accelerate our progress toward becoming carbon negative, water positive, and
zero waste, as well as to protect and preserve ecosystems.
When it comes to our carbon footprint, we continue to support clean electricity infrastructure
through long-term investments to bring more power onto the grids where we operate. Since
setting our carbon negative target, we have contracted over 34 gigawatts of renewable
energy, including projects in 24 countries.
In our efforts to become water positive, we provided more than 1.5 million people with
access to clean water and sanitation, achieving our water access target.
Our journey to zero waste includes reducing waste at our campuses and datacenters and
advancing circularity for cloud hardware, packaging, and devices. Our most recent report
showed we had a reuse and recycle rate of 89.4 percent for servers and components across
all cloud hardware.
Finally, AI gives new opportunity for ecosystem and biodiversity management with solutions
like AI-assisted bioacoustics. And we’re expanding collaboration with local communities to
build and operate datacenters in ways that address local challenges and create greater
benefits. This work is guided by our Sustainability Standards for new construction and our
Datacenter Community Pledge.
Learn more about our progress and learnings as we pursue our commitments in our annual
Impact Summary.
OUR CULTURE
Just as our culture has been critical in getting us to this point, it will be critical to our success
going forward. At Microsoft, we think of our culture as being both input and output. To pursue
new concepts, we need new capability. To build new capability, we need a culture that allows
us to grow that capability long before it is conventional wisdom. For us, that means
constantly exercising our growth mindset and confronting our fixed mindset—each one of us,
every day. It is the only way we will succeed.
Our growth mindset culture helps us in our continuous pursuit of high performance. It
doesn’t matter what we said about our culture 10 years ago or even last year if we aren’t
practicing it today—by anticipating the unmet and unarticulated needs of customers; by
working together as One Microsoft to deliver the best end-to-end solutions and services; and
by actively seeking diversity and embracing inclusion—to ensure our workforce represents
the planet we serve and the products we build always meet our customers’ needs. In our
latest Diversity & Inclusion Report, we share the ways our longstanding commitment to
diversity and inclusion endures, and what we’re learning as we continue to hire, develop,
and grow a global workforce that best supports each other and our customers. This is how
we thrive—as individuals, as teams, and as an organization. And, when we thrive, we can
help our customers and the world thrive too.
Giving also remains core to our culture. This year, more than 106,000 employees gave $250
million (including company match) to nearly 35,000 nonprofits in 111 countries. And our
employees volunteered over 1 million hours to causes they care about. I am deeply grateful
for my colleagues’ dedication to making a difference. Together, we can continue to empower
everyone around the world.
**
In closing, this is a consequential time for our company, our industry, and the world.
Ultimately, our mission requires that we translate technology into empowerment for
everyone, into real-world impact. At the end of the day, that’s what really matters.
Nearly two years later, I can’t stop thinking about the Indian farmer I met in January 2023.
He was able to apply for complex government farm subsidies using just his voice, thanks to
an app built with GPT 3.5. It was remarkable. A frontier model developed on the West Coast
of the US just months earlier was being used to directly improve the lives of rural farmers on
the other side of the globe.
That rate of diffusion was unlike anything I had seen in my career. And the pace has only
increased. Earlier this year, I was in Thailand, where I met developers using Phi-3 to optimize
their operations just days after the small language model was released.
To me, that represents the true democratization of expertise. Where the internet era put
information at our fingertips, AI is putting expertise at our fingertips. Impact like this is why
we are in this industry, and it is what gives all of us at Microsoft deep meaning in our work.
It is why we are investing in our fundamentals, in our people, and in continued innovation—
so that we can help others achieve more for the long term.
It is not an exaggeration to say that what each of us does right now with the unique
opportunity we have been given will shape the future. And I look forward to seeing how all of
us use Microsoft as a platform to make a difference—one customer, one community, one
country at a time.
Satya Nadella
Chairman and Chief Executive Officer
October 18, 2024