Andy Grove: The Hungarian Refugee Who Made Intel Paranoid Enough to Win

Santa Clara, California. The middle of 1985. 2 men sit in an office at Intel. One of them is Gordon Moore, the chairman, the scientist whose name is already attached to a law. The other is Andy Grove, the president. He is 48 years old. He speaks English with a Hungarian accent. And he does not like wasting words. The company they built is bleeding. For more than a decade, Intel has made its living on memory chips. Now Japanese manufacturers are flooding the market with memory at prices Intel cannot match. Every quarter, the numbers get worse. Meeting after meeting ends with no decision. Grove later wrote about this moment in his book, Only the paranoid survive. He looked out the window at a Ferris wheel turning in the distance at an amusement park. Then he turned back to Moore and asked a question. If we got kicked out and the board brought in a new CEO, what do you think he would do? Moore did not hesitate. He would get us out of memories. Grove stared at him. Then he said the line that would change the company. Why shouldn't you and I walk out the door, come back in, and do it ourselves? It sounds simple. It was not. Memory was Intel's identity. It was the business the company was founded to win. Walking away meant closing plants, firing thousands of people, and betting everything on a small chip that most of the world still thought of as a toy. Grove was about to do it anyway. To understand why this man could make a decision that almost no one else in his position could make, we have to go back to Budapest, to a little boy with a false name. He was born Andras Grof in Budapest, Hungary, on the 2nd of September, 1936, an only child. His parents ran a small dairy business, and by his own account, his early childhood was happy and comfortable. At four years old, he caught scarlet fever. It nearly killed him. It left him with damaged hearing for the rest of his life. Then the war came to Hungary. His family was Jewish. In 1942, his father was drafted into a labor battalion and sent to the Russian front to clear roads and build fortifications for the army. The next year. The family was told that he had disappeared. In 1944, when András was 8, Nazi Germany occupied Hungary. Within months, hundreds of thousands of Hungarian Jews were deported to the camps, most of them to Auschwitz. András and his mother survived by becoming other people. They took false identities and were hidden by friends. A small boy learned to answer to a name that was not his, And to keep his mouth shut. When the war ended, the Soviets arrived. They were not liberators for long. And then, in the fall of 1945, there was a figure at the door. In his memoir, Grove described his father that day as an emaciated man, filthy, in a ragged soldier's uniform. He had come home. So by the age of 9, Andy Grove had already learned 3 things. The world can turn on you overnight. The people in charge will not save you. And the only way through is to see the danger before it reaches you. Years later, people would call him paranoid. He never treated it as an insult. Communist Hungary was a country being slowly strangled, as Groves later put it. But he found something to hold on to: chemistry. He was good at it. He was a university student, one of a small group being prepared for leadership in the new Hungary. Then, in October 1956, students in Budapest rose up against the Soviet-backed government. For a few days, it looked like they had won. Then the Russian tanks came back into the city and crushed the revolution. The border with Austria was still open. but not for long. An aunt who had survived Auschwitz gave him blunt advice: Go. His parents agreed. One night he said goodbye, left his house key with his mother, and walked out. He traveled with other students toward the border, through the countryside, on foot for miles until they reached a farmhouse near the frontier. He was 20 years old. He had almost no money, and he barely spoke English. In the dark, he made the final run across. On the other side, he stumbled into a stranger. The man spoke to him in Hungarian. Relax, he said. You are in Austria. Grove never went back. Decades later, a reporter asked him if he had ever wanted to return to Hungary. Why do I want to go back to a place that did not want me? he said. Relief organizations got him across the Atlantic. In 1957, he arrived in the United States, and relatives in New York City took him in. He enrolled at the City College of New York, a free public college that had lifted generations of immigrants. He studied chemical engineering in a language he was still learning. With his damaged hearing, every lecture was work. He kept going. That first summer, he took a job as a busboy at a resort in New Hampshire. One of the waitresses was another refugee, a young woman from Austria named Eva Kastan, who was studying at Hunter College. They were married in Queens in June 1958. They stayed married for the rest of his life. He also changed his name. Andros Grof became Andrew Stephen Grove, simpler for Americans to say. A new name again. But this time, one he chose. In 1960, he graduated. The New York Times noticed. It ran a story about the Hungarian refugee who had become a senior in engineering. Then he and Eva drove west, to the University of California at Berkeley, where he earned a doctorate in chemical engineering in 1963. That same year, Grove took a job at Fairchild Semiconductor in Mountain View. The company where the modern silicon chip was being invented. Fairchild was full of brilliant people. It was also chaos. Grove became a researcher in the lab run by Gordon Moore. He was very good. He studied how silicon devices behaved and why they failed. By 1967, he was assistant director of development, and he wrote a college textbook called Physics and technology of semiconductor devices. Universities used it for years. But what he saw at Fairchild bothered him. Great ideas came out of the lab and then died somewhere between research and the factory floor. Nobody owned the handoff. Nobody was held to a number. He was a scientist who had grown up in a world without order. He wanted order. He wanted to know who was responsible for what and by when. In 1968, Moore and Robert Noyce, the co-inventor of the integrated circuit, decided to leave Fairchild and start their own company. Grove left with them. He would not be a founder. He went anyway. Intel was incorporated in Mountain View on the 18th of July, 1968. Grove joined on the day it was incorporated. He was not a founder. He held no big block of founder stock. He was the first employee they hired. His title was Director of Engineering. In practice, his job was to take the ideas of 2 famous scientists and make them come out of a factory— on time, in volume, and working. The plan was memory. Computers in 1968 stored data on tiny magnetic rings strung on wires. Noyce and Moore were betting that silicon chips would replace them. Intel's first products were memory chips, and in 1970, the company released the 1103, the first commercially available dynamic memory chip. By 1972, it was the best-selling memory chip in the world. Making them was brutal. Early yields were terrible. Most of the chips coming off the line did not work. It was Grove's problem to fix. And he fixed it the way he would fix everything for the next 30 years. He measured. He asked why. He asked who. Here is how Intel was split. Noyce was the face. The charmer, the man who could walk into any room and make it believe. Moore was the scientist. The quiet strategist who could see 10 years ahead. Grove was the operator. The one who turned belief and vision into shipments. In its first year, Intel's revenue was $2,672. That is not a typo; it was roughly the price of a used car. Grove ran Intel like a man who knew how quickly things could fall apart. In the early 1970s, he started what became known as the late list. Anyone who arrived after 8 in the morning had to sign in, With their arrival time. Security guards stood at the doors. Engineers who had been working until midnight were furious. Grove later admitted he never actually read the list. It did not matter. Its existence got people in on time. He invented a phrase for the way meetings at Intel would work: constructive confrontation. You attacked the problem, not the person. And you could attack the boss. It's give and take, and anyone in the company can yell at him, Craig Barron, who later succeeded him as chief executive, said. He's not above it. There were no executive dining rooms, no reserved parking. Grove sat in a standard cubicle, the same as everyone else. I've been living in cubicles since 1978, he said, and it hasn't hurt a whole lot. He could be harsh. Fortune called him mercurial and demanding. People who disagreed with him had to be ready because he would push back hard, and he expected the same in return. Many people loved working for him. Some did not last. But the point of the harshness was always the same: the truth had to travel up. In a company where the boss could not be told bad news, Grove believed, The bad news would arrive anyway, only later and bigger. He had a favorite way of saying it: The devil is in the details. Intel Vice President Dennis Carter remembered that Grove would even correct the spelling in his memos, in Grove's second language. One of Grove's habits outlived him by decades. He believed that a goal without a measurement was just a wish. So at Intel, every team set objectives, and under each objective, a few key results. The objective was where you wanted to go. The key results were how everyone would know if you got there. The key result has to be measurable, he explained. But at the end, you can look and without any arguments, did I do that or did I not do it? Yes, no. Simple. No judgments in it. In 1985, a young Intel employee named John Doerr sat in one of Grove's internal management courses and learned the system. Doerr later became a famous venture capitalist. He carried the method to a tiny search company called Google. He called Grove the father of OKRs, short for Objectives and Key Results. Larry Page later wrote that they had helped Google grow tenfold, many times over. In 1983, Grove put his methods into a book called High Output Management. It never became a bestseller in his lifetime the way his later book would, but it became a cult classic in Silicon Valley. Founders still pass it hand to hand. His core idea was simple enough to fit on a sticky note: a manager's output is the output of the people he or she runs. Not the meetings, not the memos— the output. He had come from a place where the system lied about everything. He built a company where the numbers had to be true. In late November 1979, an Intel district sales manager named Don Bookout sent a desperate 8-page telex to headquarters. Intel was losing. The company had a new 16-bit microprocessor, the 8086. Microprocessors were whole computers on a single chip. A technology Intel had pioneered with the 4004 in 1971. But Motorola had just launched a rival, the 68000. Engineers loved it. It was faster and easier to program. Customers were walking away from Intel. Bookout's boss forwarded the telex to Andy Grove, then Intel's president and chief operating officer. Grove did not wait. Within a week, the executive staff met. About a week after that, a task force was formed. Within 4 weeks, Grove and his team had reset the priorities of the entire company. They gave the campaign a name that left no room for doubt: Operation Crush. The objective was written down in one sentence: Establish the 8086 as the highest-performance 16-bit microprocessor family. And the key result was audacious: 2,000 design wins in a single year. A design win meant a customer committed to building a product around Intel's chip. Here was the move. Intel could not win on the chip alone. Motorola's chip was arguably better. So Intel stopped selling a chip and started selling a future. A family of products, development tools, support, a roadmap that showed customers where the 8086 would be in 5 years. Intel's sales force was sent out armed with videos of Grove himself, Spurring them on. By January 1980, Crush teams were in field offices around the world. Intel's own history says the campaign produced nearly 2,500 design wins. Motorola's chip was routed. And one of those design wins was a small, quiet project inside a company called IBM. In 1980, an Intel field sales engineer named Earl Whetstone was working the IBM account. He spotted something. A small team inside IBM was building a personal computer on a 1-year deadline, and it had permission to use outside suppliers. Whetstone pushed. The team chose the 8088, a cheaper version of the 8086. On the 12th of August, 1981, IBM launched the personal computer. It was a hit. Then came the clones, from Compaq and dozens of others, all copying the IBM design, all needing the same Intel brain. Nobody of Intel fully understood what had happened at the time. It looked like one design win among thousands. It turned out to be the most important sale in the company's history. Every PC-compatible computer in the world would now run on Intel's architecture. Grove later wrote a line about crisis that sums up those years: Bad companies are destroyed by crisis. Good companies survive them. Great companies are improved by them. But in the early '80s, Intel still thought of itself as a memory company. And memory was about to break its heart. By the early 1980s, Japanese chipmakers had become ferociously good at making memory. Their factories had higher yields. Their prices kept falling. American companies accused them of dumping, selling below cost to drive rivals out. The Japanese makers disputed it. Whatever the cause, Intel's memory business was turning into a hole that swallowed money. Intel had fought back the way great engineers do: better technology, more research. It was not enough. Grove later wrote that for a year, the company wandered. Managers kept arguing. Nobody wanted to say the obvious thing out loud. That brings us back to the office, to the Ferris wheel, and to the question: What would a new CEO do? Here is what made that question so powerful. Grove and Moore were not short of information; they were short of distance. A new boss would carry no history, No pride, no memory of the nights spent getting the first memory chips to work. Grove simply asked them both to imagine they were strangers. Then they acted. In 1985, Intel phased out of dynamic memory chips, the business it had been founded to dominate. The price was terrible. Intel laid off more than 7,000 people, close to a third of its workforce. Plants were closed. In 1986, The company lost $173 million. In his letter to shareholders, Grove called it the toughest year in Intel's history. Some of Intel's own managers had seen it coming before the top did. Grove noticed that. He later wrote that the middle managers had quietly been moving factory capacity from memory to microprocessors for years because that was where the profit was. The people closest to the market had already voted. The leaders were the last to know. It became one of the founding ideas of his management philosophy. He called it a strategic inflection point, the moment when the ground shifts so much that the old strategy can no longer win. You either change or you start to die. Getting out of memory was the retreat. What came next was the attack. In 1985, Intel introduced the 386, a 32-bit microprocessor powerful enough to change what a personal computer could do. And Grove made a second decision that was, at the time, almost unthinkable. In those days, computer makers refused to depend on a single chip supplier. Factories broke down, production stopped. So customers demanded a second source, another company licensed to make the same chip. For earlier chips, Intel had licensed its designs to rivals, including a company called Advanced Micro Devices . For the 386, Grove said no. Intel would be the only source. To make that promise credible, he made the chip in 3 different factories in 3 different states: Santa Clara, California; Hillsboro, Oregon; And Chandler, Arizona. If one plant went down, 2 would still be running. He convinced the customers that Intel alone was reliable enough. It worked. In 1986, Compaq shipped the first computer built on the 386, beating IBM itself. The PC industry exploded, and this time, the profits on the brain of every machine went to one company. There was a cost. AMD said Intel had broken their agreement. It went after Intel, and in the end, Intel paid it millions of dollars in damages. But AMD could no longer simply copy Intel's new designs. It had to build its own. Inside Intel, the company now depended so heavily on one line of products that an employee joked there was only one product, and Andy Grove was its product manager. By the end of the 1980s, Intel had a new problem. Its chips were inside millions of computers, and almost nobody who bought a computer knew it. Cheaper rivals were selling compatible chips. Computer makers used them to cut prices. To the shopper in the store, a chip was a chip. In 1989, Intel's Dennis Carter met a computer reseller in Arizona, MicroAge. Its marketing chief suggested a test. Intel would pay for a billboard telling people to make sure their computer had Intel Inside. The stores would put stickers on the Intel machines. In 1991, it became a worldwide campaign: Intel Inside. Intel paid part of the advertising costs of any computer maker that put the logo in its ads. The little swirl and the 5-note jingle became some of the most recognized branding on Earth. It was a radical idea. A parts supplier talking directly to the end customer, over the heads of the companies that actually bought its parts. Ingredient branding, marketers called it. Almost nobody had done it apart from a sweetener called NutraSweet. Under Grove, Intel was no longer just a supplier. It was a brand people asked for by name. And that is exactly why the next crisis hit so hard. In June 1994, a mathematics professor named Thomas Nicely at Lynchburg College in Virginia noticed something wrong. He was using computers to study prime numbers, and the answers coming out of his new Pentium machine did not match. He spent months ruling out other causes. On the 24th of October, he reported the problem to Intel. A week later, he emailed other academics, and the story raced across the early internet. The flaw was real. 5 missing entries in a table inside the chip meant that in rare cases, division came out slightly wrong. Byte magazine estimated about 1 in 9 billion random divisions would be affected. Intel had already found the bug itself and was fixing it in new chips. It had not told anyone. Grove's first instinct was the engineer's instinct. The math said almost nobody would ever be affected. So Intel said the flaw was not serious and offered replacements only to customers who could prove they needed one. It was a disaster. CNN ran the story, the New York Times covered it, the Boston Globe put it on the front page. Then IBM paused shipments of its own Pentium computers. Intel stock fell Sell hard. Grove was forced to see what he had missed. On the 20th of December, 1994, Intel announced it would replace any flawed Pentium for anyone who asked. In January 1995, it took a charge of $475 million. He was honest about the mistake. I think the kernel of the issue we missed, he told The Wall Street Journal, was that we presumed to tell somebody what they should or shouldn't worry about. Or should or shouldn't do. Intel Inside had made the company a household name. A household name cannot argue statistics with its customers. It has to keep their trust. And the crisis, in the end, did not sink Intel. Its sales kept growing. The company invested heavily in new methods to verify its chips mathematically before they shipped. It was another strategic inflection point, and this time, Grove saw it in weeks, not years. Winning on that scale made enemies. By the late 1990s, Intel controlled the large majority of the market for personal computer processors, and it fought hard to keep it. Critics, rivals, and regulators accused the company of aggressive and anti-competitive tactics. Intel was sued by rivals and investigated by the Federal Trade Commission. In June 1998, the FTC sued Intel. It said Intel had abused its monopoly power by cutting off technical information to 3 customers— Digital Equipment, Compaq, and Intergraph— when they were in patent disputes with it. A federal judge in a separate case, brought by Intergraph, had said in a preliminary ruling that Intel had probably abused its power. Intel denied all of it. Intel believes its actions are well within its rights under existing intellectual property and antitrust laws, the company said. It argued that it was protecting its own patents. In March 1999, the night before the trial was due to start, Intel and the FTC reached a settlement. Intel agreed not to withhold technical information from customers in such disputes, with limited exceptions. It admitted no wrongdoing. The remaining claims in Intergraph's own antitrust case were dismissed that same month. So, the record has 2 sides. To its rivals, Intel under Grove was a bully. To Intel, it was a company defending what it had invented and paid for. Both things can be true of a company that wins as completely as Intel did. Grove had become president of Intel in 1979 and chief executive in 1987. In 1995, he was diagnosed with prostate cancer. He did what he always did. He researched it like an engineering problem, read the studies, compared the data, and chose his own treatment. Then he wrote a cover story for Fortune about it, criticizing doctors for not sharing data the way chip engineers did. In 1996, he published Only the Paranoid Survive, The title came from his motto. Success breeds complacency, he wrote. Complacency breeds failure. Only the paranoid survive. In December 1997, Time magazine named him Man of the Year. Time called him the person most responsible for the amazing growth in the power and the innovative potential of microchips. The companion profile was called A Survivor's Tale. The numbers behind that cover are hard to take in. Intel's revenue went from $2,672 in its first year to $20.8 billion in 1997. While Grove was chief executive, Intel's stock market value went from about $4 billion to about $197 billion, an increase of more than 4,000%. Intel became one of the most valuable companies in the world. With 64,000 employees. The boy who had hidden under a false name now ran the company inside most of the world's computers. In May 1998, Grove handed the chief executive job to Craig Barrett. He stayed on as chairman until 2004, and then as a senior advisor. In 2000, he was diagnosed with Parkinson's disease. He gave tens of millions of dollars to research, and he pushed scientists to move with the urgency of the chip industry. He gave $26 million to the City College of New York, the school that had taken in a 20-year-old refugee with almost no English. It was the largest gift in the college's history. Its engineering school is now the Grove School of Engineering. He did not stop arguing either. In 2010, He wrote that America was wildly inefficient at creating tech jobs, that too much manufacturing had moved to Asia, and that the country needed to build at home. Many in Silicon Valley disagreed. He did not mind. He had been arguing with powerful people all his life. Andy Grove died at his home in Los Altos, California, on the 21st of March, 2016. He was 79 years old. Go back one last time to that office in 1985. 2 men. A Ferris wheel turning in the window. A company losing money on the business it loved. Most leaders in that chair do what feels loyal. They defend the past. They fight one more year. They wait for the market to come back. Grove asked what a stranger would do, and then he became the stranger. Maybe that came from Budapest, from a childhood where the world changed overnight, and the only people who survived were the ones who moved first. He always said he could not point to exactly how his past shaped the way he ran a business, but he never stopped watching for the danger Intel walked out of memory and into microprocessors. It became the brain of the personal computer and, for a generation, the most important technology company in the world. Years later, Grove called it the best decision they ever made. He also admitted it had been agonizing. The kid who crossed the border in the dark had learned a lesson early. And he never let anyone at Intel forget it. Only the paranoid survive.