The Five-Letter Word
In Kuala Lumpur, in the spring of 2024, a Malaysian agritech founder named Adrian Lee was demonstrating his company's AI chatbot — the thing helped farmers tend bok choy — when Satya Nadella spotted something that shouldn't have been there. A five-letter word, tucked inside a box, lurking in the corner of a PowerPoint slide flashed for a fraction of a second: Llama. Not the animal. The rival AI model, built by Meta, sitting right there in the tech stack of a company chosen to showcase Microsoft's technology to Microsoft's own CEO. "You're doing also Llama? You're using both?" Nadella asked, a note of surprise cutting through the convention hall noise. Lee, the CTO of Agroz, shifted. "Um, yep, so we are, we are using Llama also," he managed, embarrassment coloring his reply. Nadella didn't bristle. He didn't perform displeasure. He leaned in. "What are you using Llama for?" And when Lee explained that the compact model fit better in offline robots, Nadella pivoted — from CEO-as-chief-market-researcher to CEO-as-chief-salesman — and rattled off the parameter count of Microsoft's own small language model, Phi, from memory. Just 3.8 billion parameters. Small enough to run on the edge. Lee said Agroz would experiment with Phi. Agroz's CEO offered Nadella some company-grown bok choy. "Oh, you want me to eat it?" he said. He took a small piece. "Mmmm," he said, chewing vigorously, and no doubt chewing over the lesson about how perilously competitive the AI race actually is.
This is the texture of how Satya Nadella leads the most valuable company on Earth — not from behind a lectern, not through sweeping proclamations, but by noticing a five-letter word on someone else's slide in a secondary market and turning it into a sales call. The man who transformed a $300 billion software colossus mired in mediocrity into a $3 trillion titan operates at this granular frequency. He reads the room. He reads the slide. He reads the parameter count from memory. And then he eats the bok choy.
By the Numbers
The Nadella Era at Microsoft
~10xMarket cap growth during Nadella's tenure as CEO (from ~$300B to ~$3T+)
$13B+Total investment in OpenAI
1,000%+Microsoft stock surge from Feb 2014 through 2024
$68.7BActivision Blizzard acquisition — largest in tech history
$26.2BLinkedIn acquisition (2016)
$7.5BGitHub acquisition (2018)
32+Years at Microsoft before and during CEO tenure
The Consummate Insider's Outsider
The trajectory seems improbable only in retrospect. Satya Narayana Nadella was born on August 19, 1967, in Hyderabad, the southern Indian city that would later become home to one of Microsoft's largest research and development centers outside the United States — a coincidence that carries the faint scent of destiny, or at least of the kind of rhyming that history occasionally permits. His father, B.N. Yugandhar, was a member of the elite Indian Administrative Service, a Marxist economist and civil servant — a man of the state, trained in the rigors of public planning. His mother was a Sanskrit scholar. The household contained, in miniature, the tension between analytical rigor and humanistic inquiry that would come to define Nadella's own leadership.
His grades were terrible. His father, "very, very good academically," was, as Nadella recalled at Wharton, "pretty stunned as to how anybody could be this bad." But the elder Nadella expressed this bewilderment with a certain grace. "He told me that meant I must have some other real passion." The passion, for a time, was cricket — not computers. Team sport, not code. The lessons Nadella would later draw from cricket were real and specific: once, bowling trash for his high school team, the captain pulled him from the attack, bowled a crucial wicket himself, then returned the ball to Nadella, who went on to the best bowling spell of his young life. "It was a leadership decision he made, to not break my confidence," Nadella later told an audience at Stanford. The captain who takes over without humiliating you. The leader who restores rather than replaces. File that.
He failed his entrance exam to the Indian Institutes of Technology. This detail matters — not as the obligatory setback in the hero's journey, but because it reveals something about the distance between where Nadella started and where the world would later place him. He earned an electrical engineering degree from Mangalore University in 1988, and then did what ambitious Indian engineers in the late 1980s did: he went west. Not to Silicon Valley, not to MIT — to Milwaukee, Wisconsin, for a master's in computer science at the University of Wisconsin. He had never, by his own account, been "west of Bombay."
Milwaukee. The Rust Belt. Then Sun Microsystems — the company that, under Scott McNealy, had positioned itself as the anti-Microsoft, the champion of open systems against the creeping hegemony of Windows. Nadella was a member of its technology staff. And then, in 1992, he was lured away to join the hegemon itself. He was twenty-four years old.
Twenty-Two Years in the Machine
Consider what it means to spend twenty-two years inside a single company before becoming its CEO. Consider, especially, what it means to spend those years inside this company — a company that, during Nadella's tenure as a rising executive, became the most powerful monopoly in the history of personal computing, fought the United States Department of Justice in a landmark antitrust case, conquered the enterprise, missed the internet, missed mobile, missed social, missed search, and watched its stock flatline for more than a decade under a CEO whose signature move was screaming "Developers! Developers! Developers!" until his shirt was soaked through.
Nadella arrived to work on Windows NT, the business-class operating system that would become the backbone of corporate computing worldwide. He rose steadily: vice president of Microsoft bCentral, the small-business service, by 1999; corporate vice president of Microsoft Business Solutions by 2001; senior vice president of research and development for online services by 2007; president of the server and tools business — a $19 billion annual revenue engine — by 2011. He earned an MBA from the University of Chicago while working full-time, commuting between Redmond and Chicago, which tells you something about either his tolerance for red-eyes or his appetite for frameworks, probably both.
Steve Ballmer — the volcanic, irrepressible second CEO, who had joined Microsoft in 1980 as its thirtieth employee and who presided over a period of extraordinary revenue growth coupled with extraordinary strategic drift — reportedly warned Nadella, when he took over the nascent cloud business, that there was "no parachute" if he failed. "It was a way for him to communicate both why he cared about this business and his expectations," Nadella said later, with characteristic diplomatic precision. "He gave me the opportunity but also made clear that ultimately it's performance that matters."
The cloud business performed. As executive vice president of Microsoft's Cloud and Enterprise group, Nadella led the transformation to cloud infrastructure and services that "outperformed the market and took share from competition," in the company's own carefully worded account. Azure — the platform that would eventually become the company's second growth engine, rivaling and then chasing Amazon Web Services — grew under his watch from afterthought to strategic priority.
But twenty-two years inside any institution does something to a person. It can calcify you, turning you into a creature of process, a defender of the way things have always been done. Or it can give you a cartographer's intimacy with every corridor, every dead end, every hidden passage. Nadella's particular genius — the thing that made him, in hindsight, the only viable candidate — was that he managed to inhabit the company so thoroughly that he understood its pathologies from the inside while remaining, temperamentally, an outsider. He was the person who had been at Microsoft long enough to know where all the bodies were buried, but who somehow hadn't become one of them.
The Board Question
The selection process was baroque, as these things tend to be. The names that circulated in late 2013 and early 2014 included Ford CEO Alan Mulally, Nokia's Stephen Elop, and several internal candidates. Nadella was not the obvious choice. He was the quiet choice. The company man who had never run a P&L the size of Microsoft, who had never been a public figure, who was, as the promotional copy for his book would later put it, "mostly unknown following the brainy
Bill Gates and energetic Steve Ballmer."
A board member asked him the question that boards always ask: "Do you really want to be a CEO?"
Nadella's answer was not what the board member expected. "Only if you want me to be the CEO."
The board member pushed back. "Well, that's not how CEOs are made. CEOs want that job."
"But that's how I feel," Nadella said.
He called Ballmer about it. Ballmer — who understood the theater of corporate power because he had been performing it for fourteen years — gave him the best advice anyone gave him during the process: "Yeah, it's too late to change. You are who you are."
Yeah, it's too late to change. You are who you are.
On February 4, 2014, Satya Nadella became the third CEO in Microsoft's nearly forty-year history. Microsoft's stock closed at $36.35 that day. The company's market capitalization hovered just above $300 billion. It was, in the taxonomy of tech, a legacy company — immensely profitable, structurally important, and spiritually exhausted. Under Ballmer, it had missed the smartphone revolution, was lagging on tablets, had been thoroughly outmaneuvered by Google in search and mobile, and was even losing share in the PC operating system business that had made it a household name. Microsoft's shares had fallen more than 40% during Ballmer's fourteen-year tenure. The company employed more than 100,000 people, many of whom had come to regard internal politics as a more reliable predictor of success than product excellence.
Benedict Evans, an analyst at Andreessen Horowitz, had published a blog post titled "The Irrelevance of Microsoft." Its central argument was damning: Microsoft's share of all computing devices used to connect to the internet — PCs, phones, tablets — had plunged from 90 percent in 2009 to roughly 20 percent by 2014.
Nadella's first email to all employees ran more than a thousand words. It made no mention of Windows.
The Soul Search
"If there was any one theme I wanted to emphasize that day," Nadella later wrote in
Hit Refresh, "it was that we must discover what would be lost in the world if Microsoft just disappeared."
This is a deceptively radical question to pose to an organization of 100,000 people. Most companies, especially dominant ones, operate on the implicit assumption that the world needs them. The question Nadella asked forced the opposite exercise: prove it. And the honest answer, in February 2014, was that the world would lose something — but it would be a legacy something. Office, Windows, Exchange. Tools people used because they had always used them. Not tools that inspired devotion. Not tools that made the world feel the void.
"We had to answer for ourselves, what is the company about? Why do we exist? I told them it was time for us to rediscover our soul — what makes us unique."
The language is interesting. Not "define our strategy" or "identify our markets" — the standard MBA lexicon that Nadella, with his Chicago Booth degree, was perfectly capable of deploying. Soul. He went looking for the company's soul. And he found it, characteristically, not in the products but in the original act of creation. "After all, we were a tools company first," he told Elad Gil at a Stripe-hosted fireside chat in 2023. "Bill built the basic interpreter for the Altair. And I said, god, there's more need for tools and platforms in 2014, 2023. That's who we are, right at the core, we build tech so that others can build more tech."
This reframing — from a company that makes products to a company that makes platforms for other people's products — sounds abstract until you trace its operational consequences. It meant ending the theological war against open source. It meant putting Office on iOS and Android. It meant, eventually, acquiring GitHub for $7.5 billion in 2018 — GitHub, the platform where developers stored their code, much of it running on Linux, the open-source operating system that Ballmer had once called "a cancer." It meant releasing software under open-source licenses, contributing to open-source projects, and shipping a version of Microsoft's Teams communication app for Linux.
Aaron Levie, the co-founder and CEO of cloud storage vendor Box, which had spent its early years going directly after Microsoft products, captured the shift: "There was a little bit of a take-it-or-leave-it culture." Under Nadella, the company became "more attentive to customers' needs." By 2023, Box and Microsoft had multiple product integrations. Oracle co-founder
Larry Ellison —
Larry Ellison — visited Microsoft's Redmond headquarters for the first time that year for a joint cloud announcement. Pat Gelsinger, then CEO of VMware, said offering his company's software on Azure was akin to "a Middle East peace treaty."
Know-It-Alls and Learn-It-Alls
The cultural transformation is the part of the Nadella story that gets told most often and understood least. It is also the part that his critics find most grating — the invocation of "growth mindset" as corporate liturgy, the mandatory reading lists, the empathy-as-operating-system rhetoric. To understand it properly requires acknowledging that both the admirers and the skeptics are partially right.
The pre-Nadella culture at Microsoft was, by nearly every account, poisonous. The company's "stack ranking" system — borrowed from Jack Welch's GE — forced managers to rank employees against each other on a bell curve. Every team, regardless of performance, had to designate a bottom tier. The result was predictable: people spent more energy sabotaging colleagues than building products. Teams operated in silos. Knowledge hoarding was rational behavior. Innovation was punished if it threatened a colleague's ranking. A widely circulated cartoon depicted Microsoft's organizational chart as a series of pistols pointing at one another.
Nadella eliminated stack ranking. He distributed copies of Carol Dweck's Mindset — the Stanford psychologist's work on growth versus fixed mindsets — and made it something close to a corporate philosophy. The shift from "know-it-all" to "learn-it-all" became the defining rhetorical frame. "If you take two kids at school, one of them has more innate capability but is a know-it-all," Nadella said on the podcast Hello Monday in 2019. "The other person has less innate capability but is a learn-it-all. The learn-it-all does better than the know-it-all."
He asked his senior leadership team to read Nonviolent Communication by Marshall Rosenberg — a book about leading using compassion and understanding rather than competition and judgment. He made it clear, according to the book From Incremental to Exponential, that outdated, aggressive top-down behaviors were no longer welcome.
Kathleen Hogan, Microsoft's chief people officer, acknowledged the difficulty of the transition: making the shift was "tricky to navigate because many people wanted a simpler narrative of 'This is good; that was bad.'" The important thing, she said, was to frame it as "'This is how we have to evolve to be relevant in the future,' versus being dismissive of the past."
That distinction — evolution, not repudiation — is the key to understanding how Nadella pulled it off. A non-founder CEO who arrives and declares that everything before him was broken invites civil war. A non-founder CEO who says, essentially, we were right about who we are but wrong about how we behave — that CEO can change the culture without burning the house down. Nadella understood this instinctively, and it may be the single most important leadership decision he ever made.
If you take two kids at school, one of them has more innate capability but is a know-it-all. The other person has less innate capability but is a learn-it-all. The learn-it-all does better than the know-it-all.
The critics — and they exist, some vocally — argue that the growth mindset became its own kind of orthodoxy, a "vaguely-defined, scientifically-questionable" culture tool that could be weaponized by management against dissent. There is some truth to this. Any philosophy, when adopted by 200,000 people across dozens of countries and translated through layers of middle management, will mutate. The question is whether the mutation is worse than the disease. And by any external measure — stock price, talent acquisition, product quality, partnership ecosystem, employee satisfaction — the answer under Nadella has been no.
Bill Gurley of Benchmark captured the magnitude of the shift: "If you told me 10 years ago that a group of the smartest engineers in the land would evoke the threat, 'Do what I say or I will go to work at Microsoft,' I would not have believed you. Amazing shift in corporate reputation."
The Cloud Bet
Strategy, in Nadella's telling, follows from mission and culture. Once you know why you exist and how you behave, the what becomes, if not obvious, at least navigable. The first major strategic bet was the one Nadella had already been making before becoming CEO: cloud computing.
Microsoft was late to the cloud. Amazon Web Services had launched in 2006; by the time Nadella became CEO in 2014, AWS was already generating billions in revenue and had established itself as the default infrastructure for startups and, increasingly, enterprises. Google Cloud Platform was a serious competitor. Microsoft's Azure was a distant third.
Nadella's move was to double, then triple, then quadruple down on Azure — not as a standalone product but as the infrastructure layer beneath everything Microsoft did. Office 365, the subscription version of the Office suite, ran on Azure. Dynamics 365, the enterprise resource planning tool, ran on Azure. Xbox Live, the gaming network, ran on Azure. The strategy was elegant: rather than competing with AWS on raw infrastructure alone, Microsoft could offer enterprises a vertically integrated stack — the productivity tools they already used, the cloud they already needed, all running on the same platform. The enterprise salesforce that had spent decades building relationships with CIOs became the channel for cloud adoption.
By 2024, analysts estimated that Azure was about three-quarters the size of Amazon Web Services — up from roughly half five years earlier. The company's Intelligent Cloud segment had become a growth engine that would have been unimaginable under Ballmer.
But the cloud bet was table stakes. Every major tech company eventually figured out that cloud computing was the future. What distinguished Nadella was the second bet — the one that required not just strategic vision but a particular kind of temporal intelligence, the ability to recognize a platform shift while it was still taking shape and position the company to ride it.
The OpenAI Wager
In July 2019, Satya Nadella and
Sam Altman — then thirty-four years old, best known for running Y Combinator — sat in modest chairs against a black background, a small coffee table between them bearing two glasses of water and a tiny plant. The setup bore an uncanny resemblance to Zach Galifianakis's
Between Two Ferns. Nadella held a notecard. But he wasn't going for laughs.
Sam Altman — who had been raised in St. Louis, attended Stanford, dropped out, cofounded a company called Loopt that sold for a modest sum, and then talked his way into running Y Combinator at twenty-eight before pivoting to run OpenAI, the nonprofit-turned-for-profit AI lab he had cofounded in 2015 with, among others,
Elon Musk — explained their mission on camera: "To develop artificial general intelligence, broad AI systems that can do a lot of tasks at superhuman level. I think that this will be the most important technological development in human history."
Microsoft's initial investment was $1 billion. The tech press treated it as an interesting but speculative bet — one of many Microsoft was placing in the AI space. The company, after all, had its own substantial AI research operation. The OpenAI investment was, as Fortune described it, "a subtle acknowledgment that his company's own internal efforts to stay at the bleeding edge of AI technology were falling short."
What followed was a masterclass in escalation. Microsoft would eventually invest more than $13 billion in OpenAI, providing the startup with exclusive access to Azure's computing infrastructure while integrating OpenAI's models into its own products. GitHub Copilot, launched in 2021, used OpenAI's technology to autocomplete code. Bing, the perennial also-ran in search, got a ChatGPT-powered chatbot in February 2023 that briefly — astonishingly — made people care about Bing. Office 365 got Copilot, an AI assistant embedded in Word, Excel, PowerPoint, and Outlook. The $30-per-month AI subscription that Nadella announced in July 2023 was projected to generate an additional $14 billion in revenue, according to Macquarie Equity Research.
The partnership was codependent by design. "They depend on us to build the best systems, we depend on them to build the best models, and we go to market together," Nadella told MIT Technology Review in November 2023. The architectural logic was sound: OpenAI needed computation at a scale that only a hyperscaler could provide; Microsoft needed frontier AI models that its own research labs weren't producing fast enough. Both needed the other in a way that created genuine bilateral leverage.
Then came November 2023. OpenAI's board — a peculiar nonprofit governance structure that Nadella had accepted as a condition of the partnership — fired Sam Altman. Microsoft, despite having invested more than $10 billion, held no board seat. They were given, reportedly, one minute of notice. The following weekend was a carnival of corporate intrigue: Nadella negotiated with OpenAI's board to reinstate Altman, failed, watched OpenAI announce a new interim CEO, and then — within hours — announced that Altman and co-founder Greg Brockman, "together with colleagues," would be joining Microsoft to lead a new Advanced AI Research team. The move was so fast, so decisive, and so structurally clever that it effectively turned a crisis into leverage: Microsoft was now positioned to capture OpenAI's core talent regardless of the outcome, while also maintaining its investment in the company itself.
Altman was eventually reinstated at OpenAI. The board was restructured. The crisis passed. But Nadella's performance during those seventy-two hours — calm, opportunistic, strategically relentless — told the market everything it needed to know about the kind of CEO he had become.
When the paradigm shifts, do you have something to contribute? Because there is no God-given right to exist if you don't have anything relevant.
The Acquisition Grammar
Nadella's acquisition strategy is worth examining as a body of work, because it reveals a consistent logic that is easy to miss if you focus on the deal sizes rather than the deal structures.
The first major acquisition was one he didn't want. Microsoft's $7.2 billion purchase of Nokia's mobile-device business had been announced in 2013, before Nadella became CEO, and reportedly over his reservations. He was stuck with it. The deal closed in April 2014, and shortly afterward, Nadella announced the largest layoff in Microsoft's history: 18,000 positions eliminated, the majority involving Nokia. It was a $7.6 billion write-down within a year. Painful. Necessary. A signal that the new CEO would not throw good money after bad.
Then came the acquisitions that did reflect Nadella's logic:
Minecraft (2014, $2.5 billion): Not a productivity tool. Not an enterprise play. A game — but one with an enormous, passionate community of developers and creators, many of them children. The deal seemed odd at the time. In retrospect, it was Nadella's first move toward building Microsoft as a platform company, not just a software company. Minecraft's modding community was, in essence, a developer ecosystem.
LinkedIn (2016, $26.2 billion): The professional social network gave Microsoft something it had never had — a social graph of the global workforce. It was the most expensive acquisition in Microsoft's history at the time, and the integration strategy was unusual: rather than absorbing LinkedIn into Microsoft's product suite, Nadella left it largely autonomous. The data, however, flowed.
GitHub (2018, $7.5 billion): The repository where the world's developers stored their code. GitHub had 28 million users at the time of the acquisition. The symbolism was potent: the old Microsoft, the one that had called Linux a cancer, was now the steward of the world's largest open-source code platform. Nadella put Nat Friedman, a well-known open-source advocate, in charge.
Nuance Communications (2021, $19.7 billion): An AI and speech-recognition company with deep roots in healthcare — the domain where AI's potential to improve lives was most tangible and where regulatory moats were deepest.
Activision Blizzard (2023, $68.7 billion): The largest acquisition in the history of the technology industry. The deal faced fierce regulatory opposition in the U.S. and U.K. before finally closing. It gave Microsoft a gaming portfolio — Call of Duty, World of Warcraft, Candy Crush — that positioned it as a content powerhouse in a world where the line between gaming, entertainment, and social platforms was dissolving.
The thread connecting these deals is not obvious if you think in terms of product categories. But if you think in terms of communities — developers on GitHub, professionals on LinkedIn, gamers on Activision Blizzard's platforms, creators in Minecraft, healthcare workers using Nuance — the logic becomes clear. Each acquisition brought Microsoft closer to the people who used technology, not just the people who bought it.
The Empathy Engine
Nadella's insistence on empathy as a business capability — not a soft skill, not a nice-to-have, but a source of competitive advantage — is the part of his leadership philosophy that draws the most skepticism from hard-nosed operators and the most devotion from people who work with him. It is also the part most grounded in personal experience.
His son Zain was born with cerebral palsy and in-utero brain damage resulting in severe quadriplegia. Nadella was twenty-nine. He and his wife Anu were both only children in their families. "For multiple years, I struggled with what had happened," Nadella told an audience at Vanity Fair's New Establishment Summit in 2017. "My preconceived ideas of what the birth of my first child would be like were all out the window."
The shift, when it came, was not a revelation but a reorientation. "Then I realized I had to step up and do my duty as a father. I began to see the world through my son's eyes. That's what empathy is all about."
Nadella has consistently drawn a straight line from this personal experience to his business philosophy. "Empathy makes you a better innovator," he said at the same event. "If I look at the most successful products we have created, it comes with that ability to meet the unmet, unarticulated needs of customers." The formulation is precise: unmet, unarticulated. Not the needs customers tell you about in surveys and focus groups — those are articulated needs, and they are table stakes. The needs that no one has expressed, the problems that no one has framed as problems yet — those require empathy to identify. Design thinking, in Nadella's formulation, is empathy.
He is "deeply protective" of his thirty-minute morning run — on a treadmill, to spare his forty-nine-year-old (now fifty-seven-year-old) knees from concrete. It is, by the account of reporters who have traveled with him, one of the few moments in his "highly orchestrated day" when he is alone with his thoughts. He buys more books than he can finish, signs up for more online courses than he can complete. "I fundamentally believe that if you are not learning new things, you stop doing great and useful things," he wrote in his first email as CEO.
There is something almost monastic about this discipline — the daily run, the relentless reading, the insistence on curiosity as fuel rather than ambition — that sits in productive tension with the ferocity of his competitive instincts. The man who quotes Carol Dweck and distributes Nonviolent Communication is also the man who, spotting a rival's model name on a customer's slide in Kuala Lumpur, pivoted instantly to selling his own product.
The Refounding
Reid Hoffman — the LinkedIn co-founder who, having sold his company to Nadella, became both a beneficiary and an observer of the transformation — gave Nadella a frame that Nadella has adopted enthusiastically: the "refounding." Companies, Hoffman argued, need founders. But they also, from time to time, need re-founders — people who go back to the genesis of the company, rediscover what made it matter, and then rebuild from those foundations.
"Recognize first that I'm not a founder, obviously," Nadella told Hoffman on the
Masters of Scale podcast. "Bill and Paul founded the company. Bill and Steve built the company. So, Steve had founder status as far as I'm concerned. I felt like, 'Oh, I just can't be another CEO that comes in and says, let me try to do the same thing or incrementally build.' I felt like the company needed to be refounded."
This is a delicate claim, and Nadella makes it delicately. He is not saying he is Bill Gates. He is not saying the company was broken. He is saying something more nuanced: that the mission had been achieved ("a PC on every desk and home"), and that without a new mission, the company was operating on institutional momentum — profitable but purposeless. "In retrospect, it was easy. You could put it on a spreadsheet and calculate it and what have you, except that by the late 90s, we have more or less achieved that, at least in the developed world. And I felt ever since then a little like, okay, what's our mission? Is it mission accomplished? Is it time to return all the money to the shareholders?"
The new mission — "to empower every person and every organization on the planet to achieve more" — is deliberately capacious. It can accommodate cloud computing, AI, gaming, social networking, mixed reality, quantum computing. It is a statement about purpose, not product. And it gave Nadella the latitude to make moves — Office on iOS, GitHub, OpenAI — that would have been ideologically impossible under the old regime.
The Platform Shift
By late 2024, Nadella had steered Microsoft through two major technological transformations — from PCs to cloud, and from cloud to AI — and was in the process of navigating a third: the emergence of AI not just as a feature embedded in existing products but as a platform in its own right, one that could be as foundational as the shift from mainframe to desktop or desktop to mobile.
"Whenever you have a platform shift," he told MIT Technology Review, "the key thing is to make sure the platform is ubiquitously available for developers to build all kinds of new things." The developer obsession — the bok choy-eating tour of developer kiosks, the unannounced appearances at OpenAI DevDay and GitHub Universe, the onstage exclamation "I can code again!" — is not performative enthusiasm. It is Nadella's deepest strategic conviction: whoever wins the developers wins the platform.
This conviction connects directly to Bill Gates's original insight in the 1970s and 1980s — that the value in personal computing would accrue not to the hardware makers but to the company that provided the operating system, the platform on which everyone else's software ran. Nadella has simply transposed the insight to a new layer of the stack. In the AI era, the "operating system" is not Windows. It is the model layer — the large language models that power everything from code generation to customer service chatbots — and the infrastructure layer — the cloud computing resources those models require. Microsoft, through Azure and OpenAI, controls both.
The paranoia is real. "What keeps him up at night," Fortune's Jeremy Kahn reported, "is the possibility that his company might lose 'its license to operate' — the one that relies on maintaining public trust by being a force for good in the communities in which it operates." Nadella has lived through the antitrust case of the late 1990s. He saw what happened when a dominant company lost public trust. He noticed when Microsoft's AI push sent CO2 emissions up 30% since 2020, threatening its 2030 carbon-negative goals. He noticed when customers gave a piece of their business, however small, to a rival.
"Our industry does not respect tradition — it only respects innovation," he wrote in his first email as CEO. That sentence, a decade later, reads less like a slogan and more like a diagnosis.
The Glass and the Field
On April 4, 2025 — Microsoft's fiftieth anniversary — Nadella sat for a joint interview on CNBC with both of his predecessors. The three CEOs of Microsoft, all on camera together. Gates, the founder. Ballmer, the builder. Nadella, the refounders.
When asked about his company's relationship to the world — about tariffs, about trade, about the headwinds of nationalism in an era of global technology — Nadella's answer was characteristically precise and characteristically restrained. "As an American company, as a multinational company, I feel that we have to earn the license to operate in every country, one country at a time. Every country is going to do what is right for themselves, starting with the United States. And so, therefore, I always go back and say, is Microsoft adding local surplus? Are we making businesses in that country more productive? Are we making public sector more efficient? Are we helping with health outcomes, education outcomes? So, as long as our contribution to the local society and the local economy is there, we have a license to operate."
Earn the license to operate. One country at a time. This is the language of a man who came from somewhere, who moved to a country that was not his own on his twenty-first birthday, who spent three decades inside a company that once assumed the world owed it dominance and then learned, painfully, that it did not.
In January 2025, standing on a stage in Bangalore for the AI Tour, he addressed an Indian audience as a man who had left and made good, but who had never stopped belonging. The company he led had investments on every continent. Its market capitalization moved in tandem with national
GDP conversations. By 2026, he was reporting quarterly earnings in which the company's AI business alone was "larger than some of our biggest franchises that took decades to build."
Microsoft has a cricket field now, on its new Redmond campus. Nadella walked up to it for the first time the morning of a Build conference. "Man, it's just beautiful," he said. "On a spring day in Seattle, it feels just right."
Something about the image stays. The boy from Hyderabad who bowled trash until his captain gave him back the ball, standing on a cricket pitch built into the campus of a company that he refounded. Not the founder. Not the builder. The person who went looking for the soul and found it in the act of building platforms for other people's work. The CEO who reads the parameter count off the top of his head and eats the bok choy, who quotes Sanskrit scholars and sells cloud subscriptions, who learned empathy from his son and competition from the market and who holds both truths simultaneously, the way a bowler holds the seam — lightly, precisely, with full knowledge of which way it will swing.
