Chip War

Key Takeaways

Under Consideration — to be added.

Interconnections

Under Consideration — to be added.

Highlights

• During the crises of the Cold War, the U.S. military had used threats of brute nuclear force to defend Taiwan. Today, it relies on microelectronics and precision strikes.
• The United States still has a stranglehold on the silicon chips that gave Silicon Valley its name, though its position has weakened dangerously. China now spends more money each year importing chips than it spends on oil.
• China is devoting its best minds and billions of dollars to developing its own semiconductor technology in a bid to free itself from America’s chip choke. If Beijing succeeds, it will remake the global economy and reset the balance of military power. World War II was decided by steel and aluminum, and followed shortly thereafter by the Cold War, which was defined by atomic weapons. The rivalry between the United States and China may well be determined by computing power.
• Around a quarter of the chip industry’s revenue comes from phones; much of the price of a new phone pays for the semiconductors inside. For the past decade, each generation of iPhone has been powered by one of the world’s most advanced processor chips. In total, it takes over a dozen semiconductors to make a smartphone work, with different chips managing the battery, Bluetooth, Wi-Fi, cellular network connections, audio, the camera, and more.
• Apple designs in-house the ultra-complex processors that run an iPhone’s operating system. But the Cupertino, California, colossus can’t manufacture these chips. Nor can any company in the United States, Europe, Japan, or China. Today, Apple’s most advanced processors—which are arguably the world’s most advanced semiconductors—can only be produced by a single company in a single building, the most expensive factory in human history, which on the morning of August 18, 2020, was only a couple dozen miles off the USS Mustin’s port bow.
• Fabricating and miniaturizing semiconductors has been the greatest engineering challenge of our time. Today, no firm fabricates chips with more precision than the Taiwan Semiconductor Manufacturing Company, better known as TSMC.
• Last year, the chip industry produced more transistors than the combined quantity of all goods produced by all other companies, in all other industries, in all human history. Nothing else comes close.
• It was only sixty years ago that the number of transistors on a cutting-edge chip wasn’t 11.8 billion, but 4.
• The making of Moore’s Law is as much a story of manufacturing experts, supply chain specialists, and marketing managers as it is about physicists or electrical engineers.
• Once the chip industry took shape, it proved impossible to dislodge from Silicon Valley. Today’s semiconductor supply chain requires components from many cities and countries, but almost every chip made still has a Silicon Valley connection or is produced with tools designed and built in California. America’s vast reserve of scientific expertise, nurtured by government research funding and strengthened by the ability to poach the best scientists from other countries, has provided the core knowledge driving technological advances forward.
• Asian governments, in Taiwan, South Korea, and Japan, have elbowed their way into the chip industry by subsidizing firms, funding training programs, keeping their exchange rates undervalued, and imposing tariffs on imported chips. This strategy has yielded certain capabilities that no other countries can replicate—but they’ve achieved what they have in partnership with Silicon Valley, continuing to rely fundamentally on U.S. tools, software, and customers. Meanwhile, America’s most successful chip firms have built supply chains that stretch across the world, driving down costs and producing the expertise that has made Moore’s Law possible.
• Most of the world’s GDP is produced with devices that rely on semiconductors. For a product that didn’t exist seventy-five years ago, this is an extraordinary ascent.
• Then, over 2021, a series of accidents—a fire in a Japanese semiconductor facility; ice storms in Texas, a center of U.S. chipmaking; and a new round of COVID lockdowns in Malaysia, where many chips are assembled and tested—intensified these disruptions. Suddenly, many industries far from Silicon Valley faced debilitating chip shortages. Big carmakers from Toyota to General Motors had to shut factories for weeks because they couldn’t acquire the semiconductors they needed. Shortages of even the simplest chips caused factory closures on the opposite side of the world. It seemed like a perfect image of globalization gone wrong.
• The design is carved into silicon using some of the world’s most precise machinery, which can etch, deposit, and measure layers of materials a few atoms thick. These tools are produced primarily by five companies, one Dutch, one Japanese, and three Californian, without which advanced chips are basically impossible to make. Then the chip is packaged and tested, often in Southeast Asia, before being sent to China for assembly into a phone or computer.
• Unlike oil, which can be bought from many countries, our production of computing power depends fundamentally on a series of choke points: tools, chemicals, and software that often are produced by a handful of companies—and sometimes only by one. No other facet of the economy is so dependent on so few firms.
• Chips from Taiwan provide 37 percent of the world’s new computing power each year. Two Korean companies produce 44 percent of the world’s memory chips. The Dutch company ASML builds 100 percent of the world’s extreme ultraviolet lithography machines, without which cutting-edge chips are simply impossible to make. OPEC’s 40 percent share of world oil production looks unimpressive by comparison.
• Some foreign policy strategists in Beijing and Washington dream of decoupling the two countries’ tech sectors, but the ultra-efficient international network of chip designers, chemical suppliers, and machine-tool makers that people like Chang helped build can’t be easily unwound.
• Endless tank columns; waves of airplanes; thousands of tons of bombs dropped from the skies; convoys of ships delivering trucks, combat vehicles, petroleum products, locomotives, rail cars, artillery, ammunition, coal, and steel—World War II was a conflict of industrial attrition. The United States wanted it that way: an industrial war was a struggle America would win. In Washington, the economists at the War Production Board measured success in terms of copper and iron, rubber and oil, aluminum and tin as America converted manufacturing might into military power.
  • #Anduril
• This was a leap forward in computing—or it would have been, if not for the moths. Because vacuum tubes glowed like lightbulbs, they attracted insects, requiring regular “debugging” by their engineers.
  • Origin story?
• When Bell Labs held a press conference in June 1948 to announce that its scientists had invented the transistor, it wasn’t easy to understand why these wired blocks of germanium merited a special announcement. The New York Times buried the story on page 46. Time magazine did better, reporting the invention under the headline “Little Brain Cell.” Yet even Shockley, who never underestimated his own importance, couldn’t have imagined that soon thousands, millions, and billions of these transistors would be employed at microscopic scale to replace human brains in the task of computing.
• Kilby called his invention an “integrated circuit,” but it became known colloquially as a “chip,” because each integrated circuit was made from a piece of silicon “chipped” off a circular silicon wafer.
• Shockley had a knack for spotting talent, but he was an awful manager. He thrived on controversy and created a toxic atmosphere that alienated the bright young engineers he’d assembled. So these eight engineers left Shockley Semiconductor and decided to found their own company, Fairchild Semiconductor,
• Like Kilby, Noyce had produced an integrated circuit: multiple electric components on a single piece of semiconductor material. However, Noyce’s version had no freestanding wires at all. The transistors were built into a single block of material. Soon, the “integrated circuits” that Kilby and Noyce had developed would become known as “semiconductors” or, more simply, “chips.”
• Across America, the Soviet space program caused a crisis of confidence. Control of the cosmos would have serious military ramifications. The U.S. thought it was the world’s science superpower, but now it seemed to have fallen behind. Washington launched a crash program to catch up with the Soviets’ rocket and missile programs, and President John F. Kennedy declared the U.S. would send a man to the moon. Bob Noyce suddenly had a market for his integrated circuits: rockets.
• One MIT engineer calculated that to meet the needs of the Apollo mission, a computer would need to be the size of a refrigerator and would consume more electricity than the entire Apollo spacecraft was expected to produce.
• Fairchild was a brand-new company, run by a group of thirty-year-old engineers with no track record, but their chips were reliable and arrived on time.
• The computer that eventually took Apollo 11 to the moon weighed seventy pounds and took up about one cubic foot of space, a thousand times less than the University of Pennsylvania’s ENIAC computer that had calculated artillery trajectories during World War II.
• By 1964, Noyce bragged, the integrated circuits in Apollo computers had run for 19 million hours with only two failures, one of which was caused by physical damage when a computer was being moved.
• Haggerty intuitively understood that Jack Kilby’s integrated circuit could eventually be plugged into every piece of electronics the U.S. military used. A captivating public speaker, when he preached to Texas Instruments employees about the future of electronics, Haggerty was remembered by one TI veteran as “like a messiah speaking from the mountaintop. He seemed like he could predict everything.”
• Within a year, TI’s shipments to the Air Force accounted for 60 percent of all dollars spent buying chips to date. By the end of 1964, Texas Instruments had supplied one hundred thousand integrated circuits to the Minuteman program. By 1965, 20 percent of all integrated circuits sold that year went to the Minuteman program.
• Lathrop called the process photolithography—printing with light. He produced transistors much smaller than had previously been possible, measuring only a tenth of an inch in diameter, with features as small as 0.0005 inches in height. Photolithography made it possible to imagine mass-producing tiny transistors.
• Mass production works when everything is standardized. General Motors plugged many of the same car parts into all the Chevrolets that rolled off its assembly lines. When it came to semiconductors, companies like TI lacked the tools to know whether all the components of their integrated circuits were the same. Chemicals had impurities that at the time were impossible to test. Variation in temperature and pressure caused unexpected chemical reactions. The masks through which light was projected could be contaminated by particles of dust. A single impurity could ruin an entire production run. The only method of improvement was trial and error, with TI organizing thousands of experiments to assess the impact of different temperatures, chemical combinations, and production processes. Jack Kilby spent each Saturday pacing TI’s hallways and checking on his engineers’ experiments.
• She often worked the night shift, from 11 p.m. until 8 a.m., to make sure experiments were progressing according to plan. Gathering data took days of experimentation. Then she ran regressions on the data, using her slide rule to calculate exponents and square roots, plot the results on a graph, and then interpret them—doing it all by hand. It was a slow, laborious, painful process, relying on human “computers” to crunch numbers. Yet trial and error was the only method Texas Instruments had.
• Chang spent his days tinkering with Sylvania’s production processes and his evenings studying Shockley’s Electrons and Holes in Semiconductors, the bible of early semiconductor electronics.
• Like Chang, Noyce and Moore saw no limits to the growth of the chip industry so long as they could figure out mass production.
• Grove had worked his way into a PhD program at Berkeley. He’d written Fairchild in 1962 to ask for a job interview but was told to try again later: “We like our young men to interview with us when they have finished interviewing with everybody else,” the rejection letter explained. Grove found Fairchild’s rejection letter “condescendingly disgusting,” he recalled, an early sign of the hubris that would come to define Silicon Valley.
• Bob Noyce knew that military and space programs were crucial for Fairchild’s early success, admitting in 1965 that military and space applications would use “over 95% of the circuits produced this year.”
• Noyce declined most military research contracts, estimating that Fairchild never relied on the Defense Department for more than 4 percent of its R&D budget.
• “Selling R&D to the government was like taking your venture capital and putting it into a savings account,” Noyce declared. “Venturing is venturing; you want to take the risk.”
  • #Anduril
• Noyce slashed prices, too, gambling that this would drastically expand the civilian market for chips. In the mid-1960s, Fairchild chips that previously sold for $20 were cut to $2. At times Fairchild even sold products below manufacturing cost, hoping to convince more customers to try them.
• Bob Noyce’s price cuts had paid off, opening a new market for civilian computers that would drive chip sales for decades to come. Moore later argued that Noyce’s price cuts were as big an innovation as the technology inside Fairchild’s integrated circuits.
• Fairchild, however, was still owned by an East Coast multimillionaire who paid his employees well but refused to give them stock options, viewing the idea of giving away equity as a form of “creeping socialism.” Eventually, even Noyce, one of Fairchild’s cofounders, began wondering whether he had a future at the firm. Soon everyone began looking for the exit. The reason was obvious. Alongside new scientific discoveries and new manufacturing processes, this ability to make a financial killing was the fundamental force driving forward Moore’s Law. As one of Fairchild’s employees put it in the exit questionnaire he filled out when leaving the company: “I… WANT… TO… GET… RICH.”
• America’s decision to let Soviet scientists like Trutko study semiconductors at Stanford was surprising, given U.S. fears that the Soviet Union was catching up in science and technology. Yet every country’s electronics industry was increasingly oriented toward Silicon Valley, which so totally set the standard and pace of innovation that the rest of the world had no choice but to follow—even America’s adversaries.
• A CIA report in 1959 found that America was only two to four years ahead of the Soviets in quality and quantity of transistors produced.
• Their work attracted the attention of Shokin, the bureaucrat who managed the Soviet electronics industry, and they partnered with him to convince Khrushchev that the USSR needed an entire city devoted to producing semiconductors, with its own researchers, engineers, labs, and production facilities. Even before the towns on the peninsula south of San Francisco had become known as Silicon Valley—a term that wasn’t coined until 1971—Barr and Sarant had dreamt up their own version in a Moscow suburb.
• Shokin’s “copy it” strategy was fundamentally flawed, however. Copying worked in building nuclear weapons, because the U.S. and the USSR built only tens of thousands of nukes over the entire Cold War. In the U.S., however, TI and Fairchild were already learning how to mass-produce chips. The key to scaling production was reliability, a challenge that American chipmakers like Morris Chang and Andy Grove fixated on during the 1960s. Unlike their Soviet counterparts, they could draw on the expertise of other companies making advanced optics, chemicals, purified materials, and other production machinery. If no American companies could help, Fairchild and TI could turn to Germany, France, or Britain, each of which had advanced industries of their own.
• Spying could only get Shokin and his engineers so far. Simply stealing a chip didn’t explain how it was made, just as stealing a cake can’t explain how it was baked. The recipe for chips was already extraordinarily complicated.
• Foreign exchange students studying with Shockley at Stanford could become smart physicists, but it was engineers like Andy Grove or Mary Anne Potter who knew at what temperature certain chemicals needed to be heated, or how long photoresists should be exposed to light. Every step of the process of making chips involved specialized knowledge that was rarely shared outside of a specific company. This type of know-how was often not even written down. Soviet spies were among the best in the business, but the semiconductor production process required more details and knowledge than even the most capable agent could steal.
• Thanks to the “copy it” strategy, the USSR started several years behind the U.S. in transistor technology and never caught up.
• Zelenograd might have seemed like Silicon Valley without the sunshine. It had the country’s best scientists and stolen secrets. Yet the two countries’ semiconductor systems couldn’t have been more different. Whereas Silicon Valley’s startup founders job-hopped and gained practical “on the factory floor” experience, Shokin called the shots from his ministerial desk in Moscow.
  • Central planning almost never works if you’re trying to operate at the cutting edge of anything
• Japan was nothing but an “economic power,” de Gaulle declared, huffing to an aide after the meeting that Ikeda behaved like a “transistor salesman.” But it wouldn’t be long before all the world was looking enviously at Japan, because the country’s success selling semiconductors would make it far wealthier and more powerful than de Gaulle ever imagined.
• When World War II ended, some Americans had envisioned stripping Japan of its high-tech industries as punishment for starting a brutal war. Yet within a couple years of Japan’s surrender, defense officials in Washington adopted an official policy that “a strong Japan is a better risk than a weak Japan.” Apart from a short-lived effort to shut down Japan’s research into nuclear physics, the U.S. government supported Japan’s rebirth as a technological and scientific power. The challenge was to help Japan rebuild its economy while binding it to an American-led system.
  • #Anduril - contrast with strategy of American hubris in trying to reshape Russia and China into the ally the US wants
• Sony had the benefit of cheaper wages in Japan, but its business model was ultimately about innovation, product design, and marketing. Morita’s “license it” strategy couldn’t have been more different from the “copy it” tactics of Soviet Minister Shokin. Many Japanese companies had reputations for ruthless manufacturing efficiency. Sony excelled by identifying new markets and targeting them with impressive products using Silicon Valley’s newest circuitry technology. “Our plan is to lead the public with new products rather than ask them what kind of products they want,” Morita declared. “The public does not know what is possible, but we do.”
• Throughout the 1960s, Japanese firms paid sizeable licensing fees on intellectual property, handing over 4.5 percent of all chip sales to Fairchild, 3.5 percent to Texas Instruments, and 2 percent to Western Electric. U.S. chipmakers were happy to transfer their technology because Japanese firms appeared to be years behind.
• If only TI had found a way to market its own branded devices earlier, Haggerty later lamented, TI “would have been the Sony of consumer electronics.”
• Japan’s exports of electronics—a mix of semiconductors and products that relied on them—boomed from $600 million in 1965 to $60 billion around two decades later.
• A cigar smoking, hard-driving New Yorker, Sporck was fixated on efficiency. In an industry full of brilliant scientists and technological visionaries, Sporck’s expertise was in wringing productivity out of workers and machines alike. It was only thanks to tough managers like him that the cost of computing fell in line with the schedule Gordon Moore had predicted.
• Upon arrival in California, Sporck recalled, he was surprised that the firm “had virtually no competence in the handling of labor and labor unions. I brought this competence to my new employer.” Many companies wouldn’t have described a strategy of labor relations that culminated in management getting burned in effigy as “competent.” But in Silicon Valley, unions were weak, and Sporck was committed to keeping it that way. He and his colleagues at Fairchild were “dead set” against unions, he declared.
• Wherever they looked across California, semiconductor executives like Sporck couldn’t find enough cheap workers. Fairchild scoured the U.S., eventually opening facilities in Maine—where workers had “a hatred for the labor unions,” Sporck reported—and on a Navajo reservation in New Mexico that provided tax incentives. Even in the poorest parts of America, however, labor costs were substantial. Bob Noyce had made a personal investment in a radio assembly factory in Hong Kong, the British colony just across the border from Mao Zedong’s Communist China. Wages were a tenth of the American average—around 25 cents an hour. “Why don’t you go take a look,” Noyce told Sporck, who was soon on a plane to check it out.
• “The Chinese labor, the girls working there, were exceeding everything that was ever known,” one of Sporck’s colleagues recalled. Assembly workers in Hong Kong seemed twice as fast as Americans, Fairchild executives thought, and more “willing to tolerate monotonous work,” one executive reported.
• Fairchild continued to make its silicon wafers in California but began shipping semiconductors to Hong Kong for final assembly. In 1963, its first year of operation, the Hong Kong facility assembled 120 million devices. Production quality was excellent, because low labor costs meant Fairchild could hire trained engineers to run assembly lines, which would have been prohibitively expensive in California.
  • Beginning of the end
• Managers like Sporck had no game plan for globalization. He’d just as happily have kept building factories in Maine or California had they cost the same. But Asia had millions of peasant farmers looking for factory jobs, keeping wages low and guaranteeing they’d stay low for some time. Foreign policy strategists in Washington saw ethnic Chinese workers in cities like Hong Kong, Singapore, and Penang as ripe for Mao Zedong’s Communist subversion. Sporck saw them as a capitalist’s dream. “We had union problems in Silicon Valley,” Sporck noted. “We never had any union problems in the Orient.”
• This was a complex task, but Word understood that the best weapons were “cheap and familiar,” one of his colleagues explained, guaranteeing that they could be used often in training and on the battlefield. The microelectronics had to be designed with as little complexity as possible. Every connection that had to be soldered increased the risks to reliability. The simpler the electronics, the more reliable and more power-efficient a system would be. Many defense contractors were trying to sell the Pentagon expensive missiles, but Word told his team to build weapons priced like an inexpensive family sedan. He was on the lookout for a device that was simple and easy to use, enabling it to be quickly deployed on every type of airplane, embraced by each military service, and quickly adopted by U.S. allies, too.
  • #Anduril - attritable assets
• Chang and Shepherd first visited Taiwan in 1968 as part of an Asian tour to select a location for a new chip assembly facility. The visit couldn’t have gone worse. Shepherd reacted furiously when his steak was served with soy sauce, not the way it was usually prepared in Texas. His first meeting with Taiwan’s powerful and savvy economy minister, K. T. Li, ended acrimoniously when the minister declared that intellectual property was something “imperialists used to bully less-advanced countries.”
• Taiwanese officials like K. T. Li, who’d studied nuclear physics at Cambridge and ran a steel mill before steering Taiwan’s economic development through the postwar decades, began crystallizing a strategy to integrate economically with the United States. Semiconductors were at the center of this plan. Li knew there were plenty of Taiwanese-American semiconductor engineers willing to help. In Dallas, Morris Chang urged his colleagues at TI to set up a facility in Taiwan.
• As Americans grew skeptical of military commitments in Asia, Taiwan desperately needed to diversify its connections with the United States. Americans who weren’t interested in defending Taiwan might be willing to defend Texas Instruments. The more semiconductor plants on the island, and the more economic ties with the United States, the safer Taiwan would be. In July 1968, having smoothed over relations with the Taiwanese government, TI’s board of directors approved construction of the new facility in Taiwan. By August 1969, this plant was assembling its first devices. By 1980, it had shipped its billionth unit.
• In 1977, Mark Shepherd returned to Taiwan and met again with K. T. Li, nearly a decade after their first meeting. Taiwan still faced a risk of Chinese invasion, but Shepherd told Li, “We consider this risk to be more than offset by the strength and dynamism of Taiwan’s economy. TI will stay and continue to grow in Taiwan,” he promised. The company still has facilities on the island today. Taiwan, meanwhile, has made itself an irreplaceable partner to Silicon Valley.
• Noyce and Moore abandoned Fairchild as quickly as they’d left Shockley’s startup a decade earlier, and founded Intel, which stood for Integrated Electronics. In their vision, transistors would become the cheapest product ever produced, but the world would consume trillions and trillions of them. Humans would be empowered by semiconductors while becoming fundamentally dependent on them. Even as the world was being wired to the United States, America’s internal circuitry was changing. The industrial era was ending. Expertise in etching transistors into silicon would now shape the world’s economy. Small California towns like Palo Alto and Mountain View were poised to become new centers of global power.
• Because chipmaking was a custom business, delivering specialized circuits for each device, customers didn’t think hard about software. However, Intel’s progress with memory chips—and the prospect they would become exponentially more powerful over time—meant computers would soon have the memory capacity needed to handle complex software. Hoff bet it would soon be cheaper to design a standardized logic chip that, coupled with a powerful memory chip programmed with different types of software, could compute many different things. After all, Hoff knew, no one was building memory chips more powerful than Intel’s.
• Though Gordon Moore had first graphed the exponential increase in transistor density in his famous 1965 article, Mead coined the term “Moore’s Law” to describe it.
• “In the past 200 years we have improved our ability to manufacture goods and move people by a factor of 100,” Mead calculated. “But in the last 20 years there has been an increase of 1,000,000 to 10,000,000 in the rate at which we process and retrieve information.”
• “We are really the revolutionaries in the world today,” Gordon Moore declared in 1973, “not the kids with the long hair and beards who were wrecking the schools a few years ago.”
• Strategists like Marshall knew the only answer to the Soviet quantitative advantage was to produce better quality weapons. But how? As early as 1972, Marshall wrote that the U.S. needed to take advantage of its “substantial and durable lead” in computers. “A good strategy would be to develop that lead and to shift concepts of warfare in ways that capitalize on it,” he wrote. He envisioned “rapid information gathering,” “sophisticated command and control,” and “terminal guidance” for missiles, imagining munitions that could strike targets with almost perfect accuracy. If the future of war became a contest for accuracy, Marshall wagered, the Soviets would fall behind.
  • #Anduril - office of net assessment guy
• Individual guided munitions were a powerful innovation, but they’d be even more impactful if they could share information. Perry commissioned a special program, run via the Pentagon’s Defense Advanced Research Projects Agency (DARPA), to see what would happen if all these new sensors, guided weapons, and communications devices were integrated. Called “Assault Breaker,” it envisioned an aerial radar that could identify enemy targets and provide location information to a ground-based processing center, which would fuse the radar details with information from other sensors. Ground-based missiles would communicate with the aerial radar guiding them toward the target. On final descent, the missiles would release submunitions that would individually home in on their targets. Guided weapons were giving way to a vision of automated war, with computing power distributed to individual systems in a way never before imaginable. This was only possible because the U.S. was on track “to increase the density of chips ten to a hundredfold,” as Perry told an interviewer in 1981, promising comparable increases in computing power. “We will be able to put computers, which only ten years ago would have filled up this entire room, on a chip” and field “ ‘smart’ weapons at all levels.”
  • #Anduril - sensor fusion
• Moreover, unlike when integrated circuits were first invented, the chip industry had become less focused on military production. Firms like Intel targeted corporate computers and consumer goods, not missiles. Only consumer markets had the volume to fund the vast R&D programs that Moore’s Law required. In the early 1960s, it had been possible to claim the Pentagon had created Silicon Valley. In the decade since, the tables had turned. The U.S. military lost the war in Vietnam, but the chip industry won the peace that followed, binding the rest of Asia, from Singapore to Taiwan to Japan, more closely to the U.S. via rapidly expanding investment links and supply chains. The entire world was more tightly connected to America’s innovation infrastructure, and even adversaries like the USSR spent their time copying U.S. chips and chipmaking tools. Meanwhile, the chip industry had catalyzed an array of new weapons systems that were remaking how the U.S. military would fight future wars. American power was being recast. Now the entire nation depended on Silicon Valley’s success.
  • #Anduril
• At HP, however, Anderson didn’t simply take Toshiba and NEC seriously—he tested their chips and found that they were of far better quality than American competitors. None of the three Japanese firms reported failure rates above 0.02 percent during their first one thousand hours of use, he reported. The lowest failure rate of the three American firms was 0.09 percent—which meant four-and-a-half times as many U.S.-made chips were malfunctioning. The worst U.S. firm produced chips with 0.26 percent failure rates—over ten times as bad as the Japanese results. American DRAM chips worked the same, cost the same, but malfunctioned far more often. So why should anyone buy them?
• “We have no Dr. Noyces or Dr. Shockleys,” one Japanese journalist wrote, though the country had begun to accumulate its share of Nobel Prize winners. Yet prominent Japanese continued to downplay their country’s scientific successes, especially when speaking to American audiences. Sony’s research director, the famed physicist Makoto Kikuchi, told an American journalist that Japan had fewer geniuses than America, a country with “outstanding elites.” But America also had “a long tail” of people “with less than normal intelligence,” Kikuchi argued, explaining why Japan was better at mass manufacturing.
• The U.S. had supported Japan’s postwar transformation into a transistor salesman. U.S. occupation authorities transferred knowledge about the invention of the transistor to Japanese physicists, while policymakers in Washington ensured Japanese firms like Sony could easily sell into U.S. markets. The aim of turning Japan into a country of democratic capitalists had worked. Now some Americans were asking whether it had worked too well. The strategy of empowering Japanese businesses seemed to be undermining America’s economic and technological edge.
• Sporck had a hard-earned reputation for his ability to squeeze efficiency out of assembly line workers, but Japan’s productivity levels were far ahead of anything his workers could accomplish.
• With pride, patents, and millions of dollars at stake, the brawls between U.S. chipmakers often got personal, but there was still plenty of growth to go around. Japanese competition seemed different, however. If Hitachi, Fujitsu, Toshiba, and NEC succeeded, Sporck thought, they’d move the whole industry across the Pacific. “I worked specifically on TVs at GE,” Sporck warned. “You can drive by that facility now, it’s still empty…. We knew the dangers and we damn right well weren’t gonna let that happen to us.” Everything was at stake—jobs, fortunes, legacies, pride. “We’re at war with Japan,” Sporck insisted. “Not with guns and ammunition, but an economic war with technology, productivity, and quality.”
• Japan’s government subsidized its chipmakers, too. Unlike in the U.S., where antitrust law discouraged chip firms from collaborating, the Japanese government pushed companies to work together, launching a research consortium called the VLSI Program in 1976 with the government funding around half the budget.
  • Monopolies might actually be important to cutting edge technological progress (eg Bell Labs)
• Jerry Sanders saw Silicon Valley’s biggest disadvantage as its high cost of capital. The Japanese “pay 6 percent, maybe 7 percent, for capital. I pay 18 percent on a good day,” he complained. Building advanced manufacturing facilities was brutally expensive, so the cost of credit was hugely important. A next-generation chip emerged roughly once every two years, requiring new facilities and new machinery. In the 1980s, U.S. interest rates reached 21.5 percent as the Federal Reserve sought to fight inflation.
  • What are the economic implications of the government massively subsidizing the cost of capital for critical industries?
• Chipmakers like Hitachi and Mitsubishi were part of vast conglomerates with close links to banks that provided large, long-term loans. Even when Japanese companies were unprofitable, their banks kept them afloat by extending credit long after American lenders would have driven them to bankruptcy.
• Japanese companies had more debt than American peers but nevertheless paid lower rates to borrow.
• Yet with practically unlimited bank loans available, they could sustain losses as they waited for competitors to go bankrupt. In the early 1980s, Japanese firms invested 60 percent more than their U.S. rivals in production equipment, even though everyone in the industry faced the same cutthroat competition, with hardly anyone making much profit. Japanese chipmakers kept investing and producing, grabbing more and more market share. Because of this, five years after the 64K DRAM chip was introduced, Intel—the company that had pioneered DRAM chips a decade earlier—was left with only 1.7 percent of the global DRAM market, while Japanese competitors’ market share soared.
• In 1985, Japanese firms spent 46 percent of the world’s capital expenditure on semiconductors, compared to America’s 35 percent. By 1990, the figures were even more lopsided, with Japanese firms accounting for half the world’s investment in chipmaking facilities and equipment. Japan’s CEOs kept building new facilities so long as their banks were happy to foot the bill.
• In the two decades since physicist Jay Lathrop had first turned his microscope upside down to shine light on photoresist chemicals and “print” patterns on semiconductor wafers, the process of photolithography had become vastly more complicated.
• It often took five, ten, or twenty iterations of lithography, deposition, etching, and polishing to fabricate an integrated circuit, with the result layered like a geometric wedding cake. As transistors were miniaturized, each part of the lithography process—from the chemicals to the lenses to the lasers that perfectly aligned the silicon wafers with the light source—became even more difficult.
• As Japan’s chip industry rose, however, GCA began to lose its edge. Greenberg, the CEO, imagined himself as a business titan, but he spent less time running the business and more hobnobbing with politicians. He broke ground on a major new manufacturing facility, betting that the early 1980s semiconductor boom would continue indefinitely. Costs spun out of control. Inventory was wildly mismanaged. One employee stumbled onto a million dollars’ worth of precision lenses sitting forgotten in a closet. Stories circulated of executives buying Corvettes on company credit cards. One of Greenberg’s founding partners admitted that the company was spending money like a “drunken sailor.”
  • It wasn’t just cheap overseas labor. We could have done more to keep the processes here.
• The firm’s excesses were poorly timed. The semiconductor industry had always been ferociously cyclical, with the industry skyrocketing upward when demand was strong, and slumping back when it was not. It didn’t take a rocket scientist—and GCA had a handful on staff—to figure out that after the boom of the early 1980s, a downturn would eventually follow. Greenberg chose not to listen. “He didn’t want to hear from the marketing department that ‘there’s going to be a downturn,’ ” one employee remembered. So the company entered the mid-1980s semiconductor slump heavily overextended. Global sales of lithography equipment fell by 40 percent between 1984 and 1986. GCA’s revenue fell by over two-thirds. “If we had a competent economist on staff, we might have predicted it,” one employee remembered. “But we didn’t. We had Milt.”
• Meanwhile, GCA’s customer service atrophied. The company’s attitude, one analyst recounted, was “buy what we build and don’t bother us.” The company’s own employees admitted that “customers got fed up.” This was the…
  • Capabilities of a monopolist without the attitude
• But GCA employees admitted that, though their technology was world class, the company struggled with mass production. Precision manufacturing was essential, since lithography was now so exact that a thunderstorm rolling through could change air pressure—and thus the angle at which light refracted—enough to distort the images carved on chips. Building hundreds of steppers a year…
• In 1987, Nobel Prize−winning MIT economist Robert Solow, who pioneered the study of productivity and economic growth, argued that the chip industry suffered from an “unstable structure,” with employees job hopping between firms and companies declining to invest in their workers. Prominent economist Robert Reich lamented the “paper entrepreneurialism” in Silicon Valley, which he thought focused too much on the search for prestige and affluence rather…
• However, most of GCA’s problems were homegrown, driven by unreliable equipment and bad customer service. Academics devised elaborate theories to explain how Japan’s huge conglomerates were better at manufacturing than America’s small startups. But the mundane reality was that GCA didn’t listen to its customers, while Nikon did. Chip firms that interacted…
• Greenberg, GCA’s CEO, could never figure out how to fix the company. Up to the day he was ousted, he didn’t realize just how many of his company’s problems were internal. As he flew around the world on sales visits, drinking a Bloody Mary in first class, customers thought the firm was “shipping junk.” Employees complained that Greenberg was in hock to Wall Street, focused as much on the stock price as on the business model. To make end-of-year numbers, the company would collude with customers, shipping an empty crate with a user’s manual in December before delivering the machines themselves the subsequent year. However, it was impossible to cover up the company’s loss of market share. U.S. firms, with GCA as the leader,…
• Semiconductors are the “crude oil of the 1980s,” Jerry Sanders declared, “and the people who control the crude oil will control the electronics industry.”
• When America said that oil was a “strategic” commodity, it backed the claim with military force.
• The Defense Department recruited Jack Kilby, Bob Noyce, and other industry luminaries to prepare a report on how to revitalize America’s semiconductor industry. Noyce and Kilby spent hours at brainstorming sessions in the Washington suburbs, working with defense industrial experts and Pentagon officials.
  • Find the report
• When the U.S. had occupied Japan in the years immediately after World War II, it had written Japan’s constitution to make militarism impossible. But after the two countries had signed a mutual defense pact in 1951, the U.S. began cautiously to encourage Japanese rearmament, seeking military support against the Soviet Union. Tokyo agreed, but it capped its military spending around 1 percent of Japan’s GDP. This was intended to reassure Japan’s neighbors, who viscerally remembered the country’s wartime expansionism. However, because Japan didn’t spend heavily on arms, it had more funds to invest elsewhere. The U.S. spent five to ten times more on defense relative to the size of its economy. Japan focused on growing its economy, while America shouldered the burden of defending it.
  • #Anduril
• “We’re in a death spiral,” Bob Noyce told a reporter in 1986. “Can you name a field in which the U.S. is not falling behind?” In his more pessimistic moments, Noyce wondered whether Silicon Valley would end up like Detroit, its flagship industry withering under the impact of foreign competition.
• One of Silicon Valley’s complaints was that Japan’s government helped firms coordinate their R&D efforts and provided funds for this purpose. Many people in America’s high-tech industry thought Washington should replicate these tactics. In 1987, a group of leading chipmakers and the Defense Department created a consortium called Sematech, funded half by the industry and half by the Pentagon.
• Under Noyce’s leadership, Sematech was a strange hybrid, neither a company nor a university nor a research lab. No one knew exactly what it was supposed to do.
• After the company had invented the wafer stepper, a half decade of mismanagement and bad luck had left GCA a small player, far behind Japan’s Nikon and Canon and the Netherlands’ ASML. But when Peter Simone, GCA’s president, called Noyce to discuss whether Sematech could help GCA, Noyce told him flatly: “You’re done.”
• But GCA still didn’t have a viable business model. Being “ahead of your time” is good for scientists but not necessarily for manufacturing firms seeking sales. Customers had already gotten comfortable with equipment from competitors like Nikon, Canon, and ASML, and didn’t want to take a risk on new and unfamiliar tools from a company whose future was uncertain.
• On his first trip abroad in 1953, Morita had seen America as a country “that seemed to have everything.” He was served ice cream with a tiny paper umbrella on the top. “This is from your country,” the waiter told him, a humiliating reminder of how far behind Japan was. Three decades later, however, everything had changed. New York had seemed “glamorous” on Morita’s first visit in the 1950s. Now it was dirty, crime-ridden, and bankrupt.
• Morita’s wife Yoshiko even wrote a book explaining American dinner party customs to unfamiliar Japanese readers, titled My Thoughts on Home Entertaining. (Kimonos were discouraged; “whenever everyone wears the same kind of outfit, harmony is enhanced.”)
• Morita at first found the power and wealth represented by his American friends seductive. As America lurched from crisis to crisis, however, the aura around men like Henry Kissinger and Pete Peterson began to wane. Their country’s system wasn’t working—but Japan’s was.
• “The United States has been busy creating lawyers,” Morita lectured, while Japan has “been busier creating engineers.” Moreover, American executives were too focused on “this year’s profit,” in contrast to Japanese management, which was “long range.”
• In 1989, Morita set out his views in a collection of essays titled The Japan That Can Say No: Why Japan Will Be First Among Equals.
• “Militarily we could never defeat the United States,” Morita told an American colleague at the time, “but economically we can overcome the United States and become number one in the world.”
• What made The Japan That Can Say No truly frightening to Washington was not only that it articulated a zero-sum Japanese nationalism, but that Ishihara had identified a way to coerce America. Japan didn’t need to submit to U.S. demands, Ishihara argued, because America relied on Japanese semiconductors. American military strength, he noted, required Japanese chips.
• Soon-to-be-prime-minister Kiichi Miyazawa publicly noted that cutting off Japanese electronics exports would cause “problems in the U.S. economy,” and predicted that “the Asian economic zone will outdo the North American zone.” Amid the collapse of its industries and its high-tech sector, America’s future, a Japanese professor declared, was that of “a premier agrarian power, a giant version of Denmark.”
• “I now regret my association with this project,” Morita told reporters, “because it has caused so much confusion. I don’t feel U.S. readers understand that my opinions are separate from Ishihara’s. My ‘essays’ express my opinions and his ‘essays’ express his opinions.”
• The same year that Ishihara and Morita published The Japan That Can Say No, former defense secretary Harold Brown published an article that drew much the same conclusions. “High Tech Is Foreign Policy,” Brown titled the article. If America’s high-tech position was deteriorating, its foreign policy position was at risk, too.
  • #Anduril - crazy that we’ve known this for decades and done almost nothing about it
• America’s supply chain statecraft had worked brilliantly in fending off Communists, but by the 1980s, the primary beneficiary looked to have been Japan. Its trade and foreign investment had grown massively. Tokyo’s role in Asia’s economics and politics was expanding inexorably. If Japan could so swiftly establish dominance over the chip industry, what would stop it from dethroning America’s geopolitical preeminence, too?
• The chip industry was full of PhDs, but Simplot hadn’t finished eighth grade. His expertise was potatoes, as everyone knew from the white Lincoln Town Car he drove around Boise. “Mr. Spud,” the license plate declared. Yet Simplot understood business in a way Silicon Valley’s smartest scientists didn’t. As America’s chip industry struggled to adjust to Japan’s challenge, cowboy entrepreneurs like him played a fundamental role in reversing what Bob Noyce had called a “death spiral” and executing a surprise turnaround.
• As Japanese firms grabbed market share, CEOs of America’s biggest chip firms spent more and more time in Washington, lobbying Congress and the Pentagon. They set aside their free-market beliefs the moment Japanese competition mounted, claiming the competition was unfair.
• However, as all of Silicon Valley’s tech titans were fleeing DRAM chips amid the Japanese onslaught, Simplot instinctively understood that Ward and Joe Parkinson were entering the memory market at exactly the right time. A potato farmer like him saw clearly that Japanese competition had turned DRAM chips into a commodity market. He’d been through enough harvests to know that the best time to buy a commodity business was when prices were depressed and everyone else was in liquidation. Simplot decided to back Micron with $1 million. He’d later pour in millions more.
• The allegation that Japanese firms were cutting prices by too much was a bit rich coming from Simplot. Whether spud or semiconductor, he’d always said business success required being “the lowest-cost producer of the highest-quality product.”
• Micron focused ruthlessly on costs because it had no choice. There was simply no other way for a small Idaho startup to win customers. It helped that land and electricity were cheaper in Boise than in California or in Japan, thanks in part to low-cost hydroelectric power. Survival was still a struggle. At one point, in 1981, the company’s cash balances fell so low it could cover only two weeks of payroll.
• Since the earliest days of the business, Joe Parkinson had made sure employees realized that their survival depended on efficiency, going so far as to dim hallway lights at night to save on power bills when DRAM prices fell. Employees thought he was “maniacally” focused on costs—and it showed.
• Grove described his management philosophy in his bestselling book Only the Paranoid Survive: “Fear of competition, fear of bankruptcy, fear of being wrong and fear of losing can all be powerful motivators.”
• Grove saw no other choice. “If we got kicked out and the board brought in a new CEO, what do you think he would do?”
• While waiting to see if his bet on PCs would work, Grove applied his paranoia with a ruthlessness Silicon Valley had rarely seen. Workdays started at 8 a.m. sharp and anyone who signed in late was criticized publicly. Disagreements between employees were resolved via a tactic Grove called “constructive confrontation.” His go-to management technique, quipped his deputy Craig Barrett, was “grabbing someone and slamming them over the head with a sledgehammer.”
• Intel’s new manufacturing method was called “copy exactly.” Once Intel determined that a specific set of production processes worked best, they were replicated in all other Intel facilities. Before then, engineers had prided themselves on fine-tuning Intel’s processes. Now they were asked not to think, but to replicate. “It was a huge cultural issue,” one remembered, as a freewheeling Silicon Valley style was replaced with assembly line rigor. “I was perceived as a dictator,” Barrett admitted. But “copy exactly” worked: Intel’s yields rose substantially, while its manufacturing equipment was used more efficiently, driving down costs. Each of the company’s plants began to function less like a research lab and more like a finely tuned machine.
• Grove’s restructuring of Intel was a textbook case of Silicon Valley capitalism. He recognized that the company’s business model was broken and decided to “disrupt” Intel himself by abandoning the DRAM chips it had been founded to build. The firm established a stranglehold on the market for PC chips, issuing a new generation of chip every year or two, offering smaller transistors and more processing power. Only the paranoid survive, Andy Grove believed. More than innovation or expertise, it was his paranoia that saved Intel.
• He would turn Samsung into a semiconductor superpower thanks to two influential allies: America’s chip industry and the South Korean state. A key part of Silicon Valley’s strategy to outmaneuver the Japanese was to find cheaper sources of supply in Asia. Lee decided this was a role Samsung could easily play.
• He also toured an IBM computer factory and was shocked he was allowed to take photographs. “There must be many secrets in your factory,” he told the IBM employee giving him the tour. “They can’t be replicated by mere observation,” the employee confidently responded. Replicating Silicon Valley’s success, though, was exactly what Lee planned to do.
• As in Japan, therefore, Korea’s tech companies emerged not from garages, but from massive conglomerates with access to cheap bank loans and government support.
• The U.S. didn’t simply provide a market for South Korean DRAM chips; it provided technology, too. With Silicon Valley’s DRAM producers mostly near collapse, there was little hesitation about transferring top-notch technology to Korea.
• Silicon Valley’s testosterone and stock option−fueled competition often felt less like the sterile economics described in textbooks and more like a Darwinian struggle for the survival of the fittest. Many firms failed, fortunes were lost, and tens of thousands of employees were laid off. The companies like Intel and Micron that survived did so less thanks to their engineering skills—though these were important—than their ability to capitalize on technical aptitude to make money in a hypercompetitive, unforgiving industry.
• Conway was a brilliant computer scientist, but anyone who spoke with her discovered a mind that glistened with insights from diverse fields, astronomy to anthropology to historical philosophy.
• She was shocked to find that the Valley’s chipmakers were more like artists than engineers.
• Conway and Mead eventually drew up a set of mathematical “design rules,” paving the way for computer programs to automate chip design. With Conway and Mead’s method, designers didn’t have to sketch out the location of each transistor but could draw from a library of “interchangeable parts” that their technique made possible. Mead liked to think of himself as Johannes Gutenberg, whose mechanization of book production had let writers focus on writing and printers on printing. Conway was soon invited by MIT to teach a course on this chip design methodology. Each of her students designed their own chips, then shipped the design to a fabrication facility for manufacturing. Six weeks later, having never stepped foot in a fab, Conway’s students received fully functioning chips in the mail. The Gutenberg moment had arrived.
• Despite its reputation for funding futuristic weapons systems, when it came to semiconductors DARPA focused as much on building educational infrastructure so that America had an ample supply of chip designers. DARPA also helped universities acquire advanced computers and convened workshops with industry officials and academics to discuss research problems over fine wine. Helping companies and professors keep Moore’s Law alive, DARPA reasoned, was crucial to America’s military edge.
• While pursuing his master’s degree at MIT, Jacobs studied antennas and electromagnetic theory and decided to focus his research on information theory—the study of how information can be stored and communicated.
• Government efforts were effective not when they tried to resuscitate failing firms, but when they capitalized on pre-existing American strengths, providing funding to let researchers turn smart ideas into prototype products.
• Minister Shokin’s “copy it” strategy. In 1963, the same year the USSR established Zelenograd, the city of scientists working on microelectronics, the KGB established a new division, Directorate T, which stood for teknologia. The mission: “acquire Western equipment and technology,” a CIA report warned, “and improve its ability to produce integrated circuits.”
  • #Anduril
• Stealing chip designs was only useful if they could be produced at scale in the USSR. This was difficult to do during the early Cold War but almost impossible by the 1980s. As Silicon Valley crammed more transistors onto silicon chips, building them became steadily harder. The KGB thought its campaign of theft provided Soviet semiconductor producers with extraordinary secrets, but getting a copy of a new chip didn’t guarantee Soviet engineers could produce it. The KGB began stealing semiconductor manufacturing equipment, too. The CIA claimed that the USSR had acquired nearly every facet of the semiconductor manufacturing process, including nine hundred Western machines for preparing materials needed for semiconductor fabrication; eight hundred machines for lithography and etching; and three hundred machines each for doping, packaging, and testing chips. However, a factory needed a full suite of equipment, and when machines broke down, they needed spare parts. Sometimes spare parts for foreign machines could be produced in the USSR, but this introduced new inefficiencies and defects. The system of theft and replication never worked well enough to convince Soviet military leaders they had a steady supply of quality chips, so they minimized the use of electronics and computers in military systems.
• In 1985, the CIA conducted a study of Soviet microprocessors and found that the USSR produced replicas of Intel and Motorola chips like clockwork. They were always half a decade behind.
• By traditional metrics like numbers of tanks or troops, the Soviet Union had a clear advantage in the early 1980s. Ogarkov saw things differently: quality was overtaking quantity. He was fixated on the threat posed by America’s precision weapons. Combined with better surveillance and communication tools, the ability to strike targets accurately hundreds or even thousands of miles away was producing a “military-technical revolution,” Ogarkov argued to anyone who’d listen. The days of vacuum tube−guided Sparrow missiles missing 90 percent of their targets in the skies over Vietnam were long gone.
  • #Anduril
• Planes using laser guidance for their bomb strikes hit thirteen times as many targets as comparable planes without guided munitions.
• In the years after Vietnam, the U.S. military had talked about its new capabilities, but many people didn’t take them seriously. Military leaders like General William Westmoreland, who commanded American forces in Vietnam, promised that future battlefields would be automated. But the Vietnam War had gone disastrously despite America’s wide technological advantage over the North Vietnamese. So why would more computing power change things? America’s military mostly sat in its barracks during the 1980s, except for a few small operations against third-rate opponents like Libya and Grenada. No one was sure how the Pentagon’s advanced gadgets would perform on real battlefields.
  • #Anduril
• “High-tech works,” Perry proclaimed. “What’s making all this work is weapons based on information instead of the volume of fire power,” one military analyst explained to the media. “It’s the triumph of silicon over steel,” declared a New York Times headline. “War Hero Status Possible for the Computer Chip,” read another.
  • #Anduril
• Morita had spent the previous decade lecturing Americans about their need to improve production quality, not focus on “money games” in financial markets. But as Japan’s stock market crashed, the country’s vaunted long-term thinking no longer looked so visionary. Japan’s seeming dominance had been built on an unsustainable foundation of government-backed overinvestment. Cheap capital had underwritten the construction of new semiconductor fabs, but also encouraged chipmakers to think less about profit and more about output. Japan’s biggest semiconductor firms doubled down on DRAM production even as lower cost producers like Micron and South Korea’s Samsung undercut Japanese rivals.
  • Growth may order the meal but profit will always have to pay the check.
• Japan’s own media perceived overinvestment in the semiconductor sector, with newspaper headlines warning of “reckless investment competition” and “investment they cannot stop.” CEOs of Japan’s memory chip producers couldn’t bring themselves to stop building new chip fabs, even if they weren’t profitable. “If you start worrying” about overinvestment, one Hitachi executive admitted, “you can’t sleep at night.” So long as banks kept lending, it was easier for CEOs to keep spending than to admit they had no path to profitability.
  • AI over-investment?
• Japanese DRAM makers would have benefitted from Andy Grove’s paranoia or Jack Simplot’s wisdom about commodity market volatility. Instead, they all poured investment into the same market, guaranteeing that few made much money.
• As early as 1983, Ogarkov had gone so far as to tell American journalist Les Gelb—off the record—that “the Cold War is over and you have won.”
• After over two decades with Texas Instruments, Chang had left the company in the early 1980s after being passed over for the CEO job and “put out to pasture,” he’d later say.
  • Could TI have become TSMC if he stayed?
• TI’s ultra-efficient manufacturing processes were the result of his experimentation and expertise in improving yields. The job he’d wanted at TI—CEO—would have placed him at the top of the chip industry, on par with Bob Noyce or Gordon Moore. So when the government of Taiwan called, offering to put him in charge of the island’s chip industry and providing a blank check to fund his plans, Chang found the offer intriguing. At age fifty-four, he was looking for a new challenge.
• As early as the mid-1970s, while still at TI, Chang had toyed with the idea of creating a semiconductor company that would manufacture chips designed by customers. At the time, chip firms like TI, Intel, and Motorola mostly manufactured chips they had designed in-house. Chang pitched this new business model to fellow TI executives in March 1976. “The low cost of computing power,” he explained to his TI colleagues, “will open up a wealth of applications that are not now served by semiconductors,” creating new sources of demand for chips, which would soon be used in everything from phones to cars to dishwashers. The firms that made these goods lacked the expertise to produce semiconductors, so they’d prefer to outsource fabrication to a specialist, he reasoned. Moreover, as technology advanced and transistors shrank, the cost of manufacturing equipment and R&D would rise. Only companies that produced large volumes of chips would be cost-competitive.
• TI’s other executives weren’t convinced. At the time, in 1976, there weren’t any “fabless” companies that designed chips but lacked their own fabs, though Chang predicted such companies would soon emerge. Texas Instruments was already making plenty of money, so gambling on markets that didn’t yet exist seemed risky. The idea was quietly binned.
  • Innovators dilemma
• The Taiwanese government provided 48 percent of the startup capital for TSMC, stipulating only that Chang find a foreign chip firm to provide advanced production technology. He was turned down by his former colleagues at TI and by Intel. “Morris, you’ve had a lot of good ideas in your time,” Gordon Moore told him. “This isn’t one of them.” However, Chang convinced Philips, the Dutch semiconductor company, to put up $58 million, transfer its production technology, and license intellectual property in exchange for a 27.5 percent stake in TSMC.
• From day one, TSMC wasn’t really a private business: it was a project of the Taiwanese state.
• The founding of TSMC gave all chip designers a reliable partner. Chang promised never to design chips, only to build them. TSMC didn’t compete with its customers; it succeeded if they did. A decade earlier, Carver Mead had prophesied a Gutenberg moment in chipmaking, but there was one key difference. The old German printer had tried and failed to establish a monopoly over printing. He couldn’t stop his technology from quickly spreading across Europe, benefitting authors and print shops alike.
• Were it not for Communist rule, China might have played a much larger role in the semiconductor industry. When the integrated circuit was invented, China had many of the ingredients that helped Japan, Taiwan, and South Korea attract American semiconductor investment, like a vast, low-cost workforce and a well-educated scientific elite. However, after seizing power in 1949, the Communists looked at foreign connections with suspicion. For someone like Morris Chang, returning to China after finishing his studies at Stanford would have meant certain poverty and possible imprisonment or death. Many of the best graduates from China’s universities before the revolution ended up working in Taiwan or in California, building the electronics capabilities of the PRC’s primary rivals.
• The year after China produced its first integrated circuit, Mao plunged the country into the Cultural Revolution, arguing that expertise was a source of privilege that undermined socialist equality. Mao’s partisans waged war on the country’s educational system. Thousands of scientists and experts were sent to work as farmers in destitute villages. Many others were simply killed. Chairman Mao’s “Brilliant Directive issued on July 21, 1968” insisted that “it is essential to shorten the length of schooling, revolutionize education, put proletarian politics in command…. Students should be selected from among workers and peasants with practical experience, and they should return to production after a few years study.”
• It was “reactionary,” one of Mao’s supporters argued, to see electronics as the future, when it was obvious that “only the iron and steel industry should play a leading role” in building a socialist utopia in China.
• While China’s small cadre of semiconductor engineers were hoeing China’s fields, Maoists exhorted the country’s workers that “all people must make semiconductors,” as if every member of the Chinese proletariat could forge chips at home.
• The Americans were told that Chinese scientists didn’t publish their research because they opposed “self-glorification.”
• Newspaper Guangming Ribao set the tone, calling on readers in 1985 to abandon “the formula of ‘the first machine imported, the second machine imported, and the third machine imported’ ” and replace it with “ ‘the first machine imported, the second made in China, and the third machine exported.’ ” This “Made in China” obsession was hardwired into the Communist Party’s worldview, but the country was hopelessly behind in semiconductor technology—something that neither Mao’s mass mobilization nor Deng’s diktat could easily change.
• Chang’s strategy was simple: do as TSMC had done. In Taiwan, TSMC had hired the best engineers it could find, ideally with experience at American or other advanced chip firms. TSMC bought the best tools it could afford. It focused relentlessly on training its employees in the industry’s best practices. And it took advantage of all the tax and subsidy benefits that Taiwan’s government was willing to provide.
• Unlike rivals who focused more on hiring politicians’ children than on manufacturing quality, Chang ramped up production capacity and adopted technology that was near the cutting edge. By the end of the 2000s SMIC was only a couple years behind the world’s technology leaders.
• The dominant belief in the U.S. government was that expanding trade and supply chain connections would promote peace by encouraging powers like Russia or China to focus on acquiring wealth rather than geopolitical power. Claims that the decline of America’s lithography industry would imperil security were seen as out of touch with this new era of globalization and interconnection.
• Anyone who raised the question of how the U.S. could guarantee access to EUV tools was accused of retaining a Cold War mindset in a globalizing world.
• Intel’s Otellini couldn’t have been more different from Jobs. He was hired to be a manager, not a visionary. Unlike Intel’s prior CEOs—Bob Noyce, Gordon Moore, Andy Grove, and Craig Barrett—Otellini’s background was not in engineering or physics, but in economics. He’d graduated with an MBA, not a PhD. His time as CEO saw influence shift from chemists and physicists toward managers and accountants.
• “They wanted to pay a certain price,” Otellini told journalist Alexis Madrigal after the fact, “and not a nickel more…. I couldn’t see it. It wasn’t one of these things you can make up on volume. And in hindsight, the forecasted cost was wrong and the volume was 100× what anyone thought.” Intel turned down the iPhone contract.
• Spending billions for second place was hardly appealing, especially since Intel’s PC business was still highly profitable and its data center business was growing quickly. So Intel never found a way to win a foothold in mobile devices, which today consume nearly a third of chips sold. It still hasn’t.
• The company’s leadership consistently prioritized the production of chips with the highest profit margin. This was a rational strategy—no one wants products with low profit margins—but it made it impossible to try anything new. A fixation on hitting short-term margin targets began to replace long-term technology leadership. The shift in power from engineers to managers accelerated this process.
• However, Grove was worried about the offshoring of advanced manufacturing jobs. The iPhone, which had been introduced just three years earlier, exemplified the trend. Few of the iPhone’s components were built in the U.S. Though offshoring started with low-skilled jobs, Grove didn’t think it would stop there, whether in semiconductors or any other industry. He worried about lithium batteries needed for electric vehicles, where the U.S. made up a tiny share of the market despite having invented much of the core technology. His solution: “Levy an extra tax on the product of offshored labor. If the result is a trade war, treat it like other wars—fight to win.” Many people chose to write off Grove as a representative of a bygone era. He’d built Intel a generation earlier, before the internet existed.
• Grove wasn’t convinced. “Abandoning today’s ‘commodity’ manufacturing can lock you out of tomorrow’s emerging industry,” he declared, pointing to the electric battery industry. The U.S. “lost its lead in batteries thirty years ago when it stopped making consumer electronics devices,” Grove wrote. Then it missed PC batteries, and now was far behind on batteries for electric vehicles. “I doubt they will ever catch up,” he predicted in 2010.
• Jerry Sanders, the Rolex-clad, Rolls Royce−driving brawler who founded AMD, liked to compare owning a semiconductor fab with putting a pet shark in your swimming pool. Sharks cost a lot to feed, took time and energy to maintain, and could end up killing you. Even still, Sanders was sure of one thing: he’d never give up his fabs.
• The problem was simple: each generation of technological improvement made fabs more expensive. Morris Chang had drawn a similar conclusion several decades earlier, which is why he thought TSMC’s business model was superior. A foundry like TSMC could fabricate chips for many chip designers, wringing out efficiencies from its massive production volumes that other companies would find difficult to replicate.
• By the 2000s, it was common to split the semiconductor industry into three categories. “Logic” refers to the processors that run smartphones, computers, and servers. “Memory” refers to DRAM, which provides the short-term memory computers need to operate, and flash, also called NAND, which remembers data over time. The third category of chips is more diffuse, including analog chips like sensors that convert visual or audio signals into digital data, radio frequency chips that communicate with cell phone networks, and semiconductors that manage how devices use electricity.
• For DRAM memory chips, the type of semiconductor that defined Silicon Valley’s clash with Japan in the 1980s, an advanced fab can cost $20 billion.
• Nvidia’s GPUs can render images quickly because, unlike Intel’s microprocessors or other general-purpose CPUs, they’re structured to conduct lots of simple calculations—like shading pixels—simultaneously.
• In 2006, realizing that high-speed parallel computations could be used for purposes besides computer graphics, Nvidia released CUDA, software that lets GPUs be programmed in a standard programming language, without any reference to graphics at all. Even as Nvidia was churning out top-notch graphics chips, Huang spent lavishly on this software effort, at least $10 billion, according to a company estimate in 2017, to let any programmer—not just graphics experts—work with Nvidia’s chips.
• Today Nvidia’s chips, largely manufactured by TSMC, are found in most advanced data centers. It’s a good thing the company didn’t need to build its own fab. At the startup stage, it would probably have been impossible to raise the necessary sums. Giving a couple million dollars to chip designers working in a Denny’s was already a gamble. Betting over a hundred million dollars—the cost of a new fab at the time—would have been a stretch even for Silicon Valley’s most adventurous investors. Moreover, as Jerry Sanders noted, running a fab well is expensive and time-consuming. It’s hard enough simply to design top-notch chips, as Nvidia did. If it had also had to manage its own manufacturing processes, it probably wouldn’t have had the resources or the bandwidth to plow money into building a software ecosystem.
• Qualcomm has made hundreds of billions of dollars selling chips and licensing intellectual property. But it hasn’t fabricated any chips: they’re all designed in-house but fabricated by companies like Samsung or TSMC.
• To many people in Silicon Valley, Sanders’s romantic attachment to fabs seemed as out of touch as his macho swagger. The new class of CEOs who took over America’s semiconductor firms in the 2000s and 2010s tended to speak the language of MBAs as well as PhDs, chatting casually about capex and margins with Wall Street analysts on quarterly earnings calls. By most measures this new generation of executive talent was far more professional than the chemists and physicists who’d built Silicon Valley. But they often seemed stale in comparison to the giants who preceded them. An era of wild wagers on impossible technologies was being superseded by something more organized, professionalized, and rationalized. Bet-the-house gambles were replaced by calculated risk management. It was hard to escape the sense that something was lost in the process.
• Moreover, Chang realized as early as anyone how smartphones would transform computing—and therefore how they would change the chip industry, too. The media focused on young tech tycoons like Facebook’s Mark Zuckerberg, but seventy-seven-year-old Chang had a perspective that few could match. Mobile devices would be a “game-changer” for the chip industry, he told Forbes, perceiving them as heralding shifts as significant as the PC had brought. He was committed to winning the lion’s share of this business, whatever the cost.
• “TSMC knows it is important to use everyone’s innovation,” Chang declared, “ours, that of the equipment makers, of our customers, and of the IP providers. That’s the power of the Grand Alliance.”
• Chang saw Tsai’s cost cutting as defeatist. “There was very, very little investment,” Chang told journalists afterward. “I had always thought that the company was capable of more…. It didn’t happen. There was stagnation.” So Chang fired his successor and retook direct control of TSMC.
• The company’s stock price fell that day, as investors worried he’d launch a risky spending program with uncertain returns. Chang thought the real risk was accepting the status quo. He wasn’t about to let a financial crisis threaten TSMC in the race for industry leadership.
  • Finance Weenies
• Since his earliest days at Apple, Steve Jobs had thought deeply about the relationship between software and hardware. In 1980, when his hair nearly reached his shoulders and his mustache covered his upper lip, Jobs gave a lecture that asked, “What is software?” “The only thing I can think of,” he answered, “is software is something that is changing too rapidly, or you don’t exactly know what you want yet, or you didn’t have time to get it into hardware.”
• Today, no company besides TSMC has the skill or the production capacity to build the chips Apple needs. So the text etched onto the back of each iPhone—“Designed by Apple in California. Assembled in China”—is highly misleading. The iPhone’s most irreplaceable components are indeed designed in California and assembled in China. But they can only be made in Taiwan.
• It took Trumpf a decade to master these challenges and produce lasers with sufficient power and reliability. Each one required exactly 457,329 component parts.
• ASML itself only produced 15 percent of an EUV tool’s components, he estimated, buying the rest from other firms. This let it access the world’s most finely engineered goods, but it also required constant surveillance.
• “If you don’t behave, we’re going to buy you,” ASML’s CEO Peter Wennink told one supplier. It wasn’t a joke: ASML ended up buying several suppliers, including Cymer, after concluding it could better manage them itself.
• EUV tools work in part because their software works. ASML uses predictive maintenance algorithms to guess when components need to be replaced before they break, for example. It also uses software for a process called computational lithography to print patterns more exactly.
• ASML’s EUV lithography tool is the most expensive mass-produced machine tool in history, so complex it’s impossible to use without extensive training from ASML personnel, who remain on-site for the tool’s entire life span. Each EUV scanner has an ASML logo on its side. But ASML’s expertise, the company readily admits, was its ability to orchestrate a far-flung network of optics experts, software designers, laser companies, and many others whose capabilities were needed to make the dream of EUV a reality.
• It’s easy to lament the offshoring of manufacturing, as Andy Grove did during the final years of his life. That a Dutch company, ASML, had commercialized a technology pioneered in America’s National Labs and largely funded by Intel would undoubtedly have rankled America’s economic nationalists, had any been aware of the history of lithography or of EUV technology. Yet ASML’s EUV tools weren’t really Dutch, though they were largely assembled in the Netherlands. Crucial components came from Cymer in California and Zeiss and Trumpf in Germany. And even these German firms relied on critical pieces of U.S.-produced equipment. The point is that, rather than a single country being able to claim pride of ownership regarding these miraculous tools, they are the product of many countries. A tool with hundreds of thousands of parts has many fathers.
• Having worked in Texas and California as well as in Taiwan, Chiang was always struck by the ambition and the work ethic that drove TSMC. The ambition stemmed from Morris Chang’s vision of world-beating technology, evident in his willingness to spend huge sums expanding TSMC’s R&D team from 120 people in 1997 to 7,000 in 2013.
• This hunger permeated the entire company. “People worked so much harder in Taiwan,” Chiang explained. Because manufacturing tools account for much of the cost of an advanced fab, keeping the equipment operating is crucial for profitability. In the U.S., Chiang said, if something broke at 1 a.m., the engineer would fix it the next morning. At TSMC, they’d fix it by 2 a.m. “They do not complain,” he explained, and “their spouse does not complain” either.
• Measured by thousands of wafers per month, the industry standard, TSMC had a capacity of 1.8 million while Samsung had 2.5 million. GlobalFoundries had only 700,000.
• The PC market was stagnant, because it seemed nearly everyone already had a PC, but it remained remarkably profitable for Intel, providing billions of dollars a year that could be reinvested into R&D. The company spent over $10 billion a year on R&D throughout the 2010s, four times as much as TSMC and three times more than the entire budget of DARPA. Only a couple of companies in the world spent more.
• Intel remains enormously profitable today. It’s still America’s biggest and most advanced chipmaker. However, its future is more in doubt than at any point since Grove’s decision in the 1980s to abandon memory and bet everything on microprocessors. It still has a shot at regaining its leadership position over the next half decade, but it could just as easily end up defunct. What’s at stake isn’t simply one company, but the future of America’s chip fabrication industry. Without Intel, there won’t be a single U.S. company—or a single facility outside of Taiwan or South Korea—capable of manufacturing cutting-edge processors.
• Intel stuck stubbornly to its integrated model—combining semiconductor design and manufacturing in one company—which executives there thought was still the best way to churn out chips. The company’s design and manufacturing processes were optimized for each other, Intel’s leaders argued. TSMC, by contrast, had no choice but to adopt generic manufacturing processes that could work just as well for a Qualcomm smartphone processor as an AMD server chip. Intel was right to perceive some benefits of an integrated model, but there were substantial downsides. Because TSMC manufactures chips for many different companies, it now fabricates nearly three times as many silicon wafers per year as Intel, so it has more chance to hone its process.
  • At the end of the day it always comes back to reps
• Intel’s near-monopoly in sales of processors for data centers is ending. Losing this dominant position would have been less problematic if Intel had found new markets. However, the company’s foray into the foundry business in the mid-2010s, where it tried to compete head-on with TSMC, was a flop. Intel tried opening its manufacturing lines to any customers looking for chipmaking services, quietly admitting that the model of integrated design and manufacturing wasn’t nearly as successful as Intel’s executives claimed. The company had all the ingredients to become a major foundry player, including advanced technology and massive production capacity, but succeeding would have required a major cultural change. TSMC was open with intellectual property, but Intel was closed off and secretive. TSMC was service-oriented, while Intel thought customers should follow its own rules. TSMC didn’t compete with its customers, since it didn’t design any chips. Intel was the industry giant whose chips competed with almost everyone.
• “Without cybersecurity there is no national security,” declared Xi Jinping, general secretary of the Chinese Communist Party, in 2014, “and without informatization, there is no modernization.”
  • #Anduril
• AI Superpowers, according to a widely discussed book by Kai-Fu Lee, former head of Google China.
  • #books-to-read
• Across the entire semiconductor supply chain, aggregating the impact of chip design, intellectual property, tools, fabrication, and other steps, Chinese firms have a 6 percent market share, compared to America’s 39 percent, South Korea’s 16 percent, or Taiwan’s 12 percent, according to the Georgetown researchers.
• Huahong and Grace, two other Chinese foundries, won little market share, in large part because the state-owned firms and municipal governments that controlled them meddled incessantly in business decisions.
• It would be impossible to replicate TSMC’s investment plans with private-sector funding alone. Only a government could take such a gamble. The amount of money China has put into chip subsidies and “investments” is hard to calculate, since much of the spending is done by local governments and opaque state-owned banks, but it’s widely thought to measure in the tens of billions of dollars.
• China’s import of chips—$260 billion in 2017, the year of Xi’s Davos debut—was far larger than Saudi Arabia’s export of oil or Germany’s export of cars. China spends more money buying chips each year than the entire global trade in aircraft. No product is more central to international trade than semiconductors.
• When companies are losing market share or in need of financing, moreover, they don’t have the luxury of focusing on the long term. This gives China powerful levers to induce foreign chip firms to transfer technology, open production facilities, or license intellectual property, even when foreign companies realize they’re helping develop competitors.
• Lee built Samsung from a trader of dried fish into a tech company churning out some of the world’s most advanced processor and memory chips by relying on three strategies. First, assiduously cultivate political relationships to garner favorable regulation and cheap capital. Second, identify products pioneered in the West and Japan and learn to build them at equivalent quality and lower cost. Third, globalize relentlessly, not only to seek new customers but also to learn by competing with the world’s best companies. Executing these strategies made Samsung one of the world’s biggest companies, achieving revenues equivalent to 10 percent of South Korea’s entire GDP.
• Theft of intellectual property may well have benefitted the company, but it can’t explain its success. No quantity of intellectual property or trade secrets is enough to build a business as big as Huawei. The company has developed efficient manufacturing processes that have driven down costs and built products that customers see as high-quality. Huawei’s spending on R&D, meanwhile, is world leading. The company spends several times more on R&D than other Chinese tech firms.
• A Wall Street Journal review of total subsidies provided by the Chinese government reached a figure of $75 billion, in the form of subsidized land, state-backed credit, and tax deductions at a scale far above what most Western companies get from their governments, though the benefits provided to Huawei might not be too different from what other East Asian governments provide to priority companies.
• The world has far more sensory information than our current ability to digitize, communicate, and process.
• Tesla’s cult following and soaring stock price have attracted plenty of attention, but what’s less noticed is that Tesla is also a leading chip designer. The company hired star semiconductor designers like Jim Keller to build a chip specialized for its automated driving needs, which is fabricated using leading-edge technology.
• From swarms of autonomous drones to invisible battles in cyberspace and across the electromagnetic spectrum, the future of war will be defined by computing power. The U.S. military is no longer the unchallenged leader. Long gone are the days when the U.S. had unrivaled access to the world’s seas and airspace, guaranteed by precision missiles and all-seeing sensors. The shock waves that reverberated around the world’s defense ministries after the 1991 Persian Gulf War—and the fear that the surgical strikes that had defanged Saddam’s army could be used against any military in the world—was felt in Beijing like a “psychological nuclear attack,” according to one account. In the thirty years since that conflict, China has poured funds into high-tech weaponry, abandoning Mao-era doctrines of waging a low-tech People’s War and embracing the idea that the fights of the future will rely on advanced sensors, communications, and computing. Now China is developing the computing infrastructure an advanced fighting force requires.
  • #Anduril
• Beijing’s aim isn’t simply to match the U.S. system-by-system, but to develop capabilities that could “offset” American advantages, taking the Pentagon’s concept from the 1970s and turning it against the United States. China has fielded an array of weapons that systematically undermine U.S. advantages. China’s precision anti-ship missiles make it extremely dangerous for U.S. surface ships to transit the Taiwan Strait in a time of war, holding American naval power at bay. New air defense systems contest America’s ability to dominate the airspace in a conflict. Long-range land attack missiles threaten the network of American military bases from Japan to Guam. China’s anti-satellite weapons threaten to disable communications and GPS networks. China’s cyberwar capabilities haven’t been tested in wartime, but the Chinese would try to bring down entire U.S. military systems. Meanwhile, in the electromagnetic spectrum, China might try to jam American communications and blind surveillance systems, leaving the U.S. military unable to see enemies or communicate with allies.
  • #Anduril
• Undergirding all these capabilities is a belief in Chinese military circles that warfare is not simply becoming “informationized” but “intelligentized”—inelegant military jargon that means applying artificial intelligence to weapons systems. Of course, computing power has been central to warfare for the past half century, though the quantity of 1s and 0s that can be harnessed to support military systems is millions of times larger than decades earlier. What’s new today is that America now has a credible challenger. The Soviet Union could match the U.S. missile-for-missile but not byte-for-byte. China thinks it can do both. The fate…
  • #Anduril
• In 2021, a group of American tech and foreign policy grandees chaired by former Google CEO Eric Schmidt released a report predicting that “China could surpass the United States as the world’s AI superpower.” Chinese leaders appear to agree. As China military expert Elsa Kania notes, the PLA has been talking about “AI weapons” for at least a decade, referring to systems that use “AI to pursue, distinguish, and destroy enemy targets automatically.” Xi…
• Georgetown University’s Ben Buchanan has noted that a “triad” of data, algorithms, and computing power are needed to harness AI. With the exception of computing power, China’s capabilities may already equal the United States’.
• It’s harder to say whether one side has an advantage when it comes to devising clever algorithms. Measured by the number of AI experts, China appears to have capabilities that are comparable to America’s. Researchers at MacroPolo, a China-focused think tank, found that 29 percent of the world’s leading researchers in artificial intelligence are from China, as opposed to 20 percent from the U.S. and 18 percent from Europe. However, a staggering share of these experts end up working in the U.S., which employs 59 percent of the world’s top AI researchers. The combination of new visa and travel restrictions plus China’s effort to retain more researchers at home may neutralize America’s historical skill at stripping geopolitical rivals of their smartest minds.
  • #Anduril
• One Chinese study has estimated that as many as 95 percent of GPUs in Chinese servers running artificial intelligence workloads are designed by Nvidia, for example. Chips from Intel, Xilinx, AMD, and others are crucially important in Chinese data centers. Even under the most optimistic projections, it will be half a decade before China can design competitive chips and the software ecosystem around them, and far longer before it can manufacture these chips domestically.
  • #Anduril
• A recent review of 343 publicly available AI-related People’s Liberation Army procurement contracts, by researchers at Georgetown University, found that less than 20 percent of the contracts involved companies that are subject to U.S. export controls. In other words, the Chinese military has had little difficulty simply buying cutting-edge U.S. chips off-the-shelf and plugging them into military systems. The Georgetown researchers found that Chinese military suppliers even advertise on their websites their use of American chips. The Chinese government’s controversial policy of “Civil Military Fusion,” an effort to apply advanced civilian technology to military systems, looks like it’s working. Absent a major change in U.S. export restrictions, the People’s Liberation Army will acquire much of the computing power it needs by simply buying it from Silicon Valley.
  • #Anduril - Civil Military Fusion
• In the mid-2010s, officials like Secretary of Defense Chuck Hagel began speaking about a need for a new “offset,” evoking the effort of Bill Perry, Harold Brown, and Andrew Marshall during the 1970s to overcome the USSR’s quantitative advantage. The U.S. faces the same basic dilemma today: China can deploy more ships and planes than the U.S., especially in theaters that matter, like the Taiwan Strait. “We will never try to match our opponents or our competitors tank for tank, plane for plane, person for person,” declared Bob Work, the former deputy defense secretary who is the intellectual godfather of this new offset, in a clear echo of the logic of the late 1970s. The U.S. military will only succeed, in other words, if it has a decisive technological advantage.
  • #Anduril
• Just as the Cold War was decided by electrons zipping around the guidance computers of American missiles, the fights of the future may be decided in the electromagnetic spectrum. The more the world’s militaries rely on electronic sensors and communication, the more they’ll have to battle for access to the spectrum space needed to send messages or to detect and track adversaries. We’ve only had a glimpse of what wartime electromagnetic spectrum operations will look like. For example, Russia has used a variety of radar and signals jammers in its war against Ukraine. The Russian government also reportedly obstructs GPS signals around President Vladimir Putin’s official travel, perhaps as a security measure. Not coincidentally, DARPA is researching alternative navigation systems that aren’t reliant on GPS signals or satellites, to enable American missiles to hit their targets even if GPS systems are down.
  • #Anduril
• In 2017, DARPA launched a new project called the Electronics Resurgence Initiative to help build the next wave of militarily relevant chip technology. In some ways, DARPA’s renewed interest in chips stems naturally from its history. It funded pioneering scholars like Caltech’s Carver Mead and catalyzed research into chip design software, new lithography techniques, and transistor structures. Yet DARPA and the U.S. government have found it harder than ever to shape the future of the chip industry. DARPA’s budget is a couple billion dollars per year, less than the R&D budgets of most of the industry’s biggest firms. Of course, DARPA spends a lot more on far-out research ideas, whereas companies like Intel and Qualcomm spend most of their money on projects that are only a couple years from fruition. However, the U.S. government in general buys a smaller share of the world’s chips than ever before. The U.S. government bought almost all the early integrated circuits that Fairchild and Texas Instruments produced in the early 1960s. By the 1970s, that number had fallen to 10−15 percent. Now it’s around 2 percent of the U.S. chip market. As a buyer of chips, Apple CEO Tim Cook has more influence on the industry than any Pentagon official today.
  • #Anduril
• In 2018, researchers discovered two fundamental errors in Intel’s widely used microprocessor architecture called Spectre and Meltdown, which enabled the copying of data such as passwords—a huge security flaw. According to the Wall Street Journal, Intel first disclosed the flaw to customers, including Chinese tech companies, before notifying the U.S. government, a fact that only intensified Pentagon officials’ concern about their declining influence over the chip industry.
• Beijing’s efforts to acquire advanced technology, the deep interconnections between the U.S. and Chinese electronics industries, and the two countries’ mutual reliance on fabrication in Taiwan all raise questions. America was already running slower. It’s now betting the future of its military on a technology over which its dominance is slipping. “This idea of pulling ahead with an offset,” argues Matt Turpin, an official who worked on the issue at the Pentagon, “is nearly impossible if the Chinese are in the car with us.”
• “Call forth the assault,” Xi Jinping declared. China’s leaders have identified their reliance on foreign chipmakers as a critical vulnerability. They’ve set out a plan to rework the world’s chip industry by buying foreign chipmakers, stealing their technology, and providing billions of dollars of subsidies to Chinese chip firms.
• China had driven U.S. solar panel manufacturing out of business. Couldn’t it do the same in semiconductors? “This massive $250 billion fund is going to bury us,” one Obama official worried, referencing the subsidies China’s central and local governments have promised to support homegrown chipmakers.
• The intelligence agencies and Justice Department unearthed more evidence of collusion between China’s government and its industries to push out American chip firms. Yet the twin pillars of American tech policy—embracing globalization and “running faster”—were deeply ingrained, not only by the industry’s lobbying, but also by Washington’s intellectual consensus. Moreover, most people in Washington barely knew what a semiconductor was. The Obama administration moved slowly on semiconductors, one person involved in the effort recalled, because many senior officials simply didn’t see chips as an important issue.
• In late 2016, six days before that year’s presidential election, Commerce Secretary Penny Pritzker gave a high-profile address in Washington on semiconductors, declaring it “imperative that semiconductor technology remains a central feature of American ingenuity and a driver of our economic growth. We cannot afford to cede our leadership.” She identified China as the central challenge, condemning “unfair trade practices and massive, non-market-based state intervention” and cited “new attempts by China to acquire companies and technology based on their government’s interest—not commercial objectives,” an accusation driven by Tsinghua Unigroup’s acquisition spree.
  • #Anduril
• America’s technological lead in fabrication, lithography, and other fields had dissipated because Washington convinced itself that companies should compete but that governments should simply provide a level playing field. A laissez-faire system works if every country agrees to it. Many governments, especially in Asia, were deeply involved in supporting their chip industries. However, U.S. officials found it easier to ignore other countries’ efforts to grab valuable chunks of the chip industry, instead choosing to parrot platitudes about free trade and open competition. Meanwhile, America’s position was eroding.
• The Obama administration considered imposing financial sanctions on ZTE, which would have severed the company’s access to the international banking system, but instead opted to punish the company in 2016 by restricting U.S. firms from selling to it. Export controls like this had previously been used mostly against military targets, to stop the transfer of technology to companies supplying components to Iran’s missile program, for example. But the Commerce Department had broad authority to prohibit the export of civilian technologies, too. ZTE was highly reliant on American components in its systems—above all, American chips. However, in March 2017, before the threatened restrictions were implemented, the company signed a plea deal with the U.S. government and paid a fine, so the export restrictions were removed before they’d taken force. Hardly anyone understood just how drastic a move it would have been to ban a major Chinese tech company from buying U.S. chips.
  • #Anduril
• One U.S. semiconductor executive wryly summed things up to a White House official: “Our fundamental problem is that our number one customer is our number one competitor.”
• This was a perfect case study of the state-backed intellectual property theft foreign companies operating in China had long complained of. The Taiwanese naturally understood why the Chinese preferred not to abide by intellectual property rules, of course. When Texas Instruments first arrived in Taiwan in the 1960s, Minister K. T. Li had sneered that “intellectual property rights are how imperialists bully backward countries.” Yet Taiwan had concluded it was better to respect intellectual property norms, especially as its companies began developing their own technologies and had their own patents to defend. Many intellectual property experts predicted that China would soon begin stealing less IP as its companies produced more sophisticated goods. However, the evidence for this thesis was mixed. Efforts by the Obama administration to cut a deal with China’s spy agencies whereby they agreed to stop providing stolen secrets to Chinese companies lasted only long enough for Americans to forget about the issue, at which point the hacking promptly restarted.
  • #Anduril
• Australian prime minister Malcolm Turnbull had at first been skeptical of an outright ban. According to Australian journalist Peter Hartcher, Turnbull bought himself a 474-page-book titled A Comprehensive Guide to 5G Security to study the topic so that he could ask better questions of his tech experts. Eventually he was convinced he had no choice but to ban the firm.
  • That’s what you want from politicians
• The debate was really about whether China should be stopped from playing an ever-larger role in the world’s tech infrastructure. Robert Hannigan, former head of the UK’s signals intelligence agency, argued that “we should accept that China will be a global tech power in the future and start managing the risk now, rather than pretending the west can sit out China’s technological rise.” Many Europeans also thought China’s technological advance was inevitable and therefore not worth trying to stop. The United States government didn’t agree.
• Its annual R&D spending now rivaled American tech giants like Microsoft, Google, and Intel. Of all China’s tech firms, it was the most successful exporter, giving it detailed knowledge of foreign markets. It not only produced hardware for cell towers, it also designed cutting-edge smartphone chips. It had become TSMC’s second biggest customer, behind only Apple. The pressing question was: Could the United States let a Chinese company like this succeed?
• On the National Security Council, however, competition with China was now seen primarily in zero-sum terms. These officials interpreted Huawei not as a commercial challenge but as a strategic one. Sony and Samsung were tech firms based in countries that were allied with the U.S. Huawei was a national champion of America’s primary geopolitical rival. Viewed through this lens, Huawei’s expansion was a threat.
• “The United States needs to strangle Huawei,” Republican senator Ben Sasse declared in 2020. “Modern wars are fought with semiconductors and we were letting Huawei use our American designs.”
• Around this time, two academics, Henry Farrell and Abraham Newman, noticed that international political and economic relations were increasingly impacted by what they called “weaponized interdependence.” Countries were more intwined than ever, they pointed out, but rather than defusing conflicts and encouraging cooperation, interdependence was creating new venues for competition. Networks that knit together nations had become a domain of conflict. In the financial sphere, the U.S. had weaponized other countries’ reliance on access to the banking system to punish Iran, for example. These academics worried that the U.S. government’s use of trade and capital flows as political weapons threatened globalization and risked dangerous unintended consequences. The Trump administration, by contrast, concluded it had unique power to weaponize semiconductor supply chains.
  • #Anduril
• They restricted any goods made with U.S.-produced technology from being sold to Huawei, too. In a chip industry full of choke points, this meant almost any chip. TSMC can’t fabricate advanced chips for Huawei without using U.S. manufacturing equipment. Huawei can’t design chips without U.S.-produced software. Even China’s most advanced foundry, SMIC, relies extensively on U.S. tools. Huawei was simply cut off from the world’s entire chipmaking infrastructure, except for chips that the U.S. Commerce Department deigned to give it a special license to buy.
• Beijing has evidently calculated that it’s better to accept that Huawei will become a second-rate technology player than to hit back against the United States. The U.S., it turns out, has escalation dominance when it comes to severing supply chains. “Weaponized interdependence,” one former senior official mused after the strike on Huawei. “It’s a beautiful thing.”
  • #Anduril
• It’s commonly argued that the escalating tech competition with the United States is like a “Sputnik moment” for China’s government. The allusion is to the United States’ fear after the launch of Sputnik in 1957 that it was falling behind its rival, driving Washington to pour funding into science and technology. China certainly faced a Sputnik-scale shock after the U.S. banned sales of chips to firms like Huawei. Dan Wang, one of the smartest analysts of China’s tech policy, has argued that American restrictions have “boosted Beijing’s quest for tech dominance” by catalyzing new government policies to support the chip industry. In the absence of America’s new export controls, he argues, Made in China 2025 would have ended up like China’s previous industrial policy efforts, with the government wasting substantial sums of money. Thanks to U.S. pressure, China’s government may provide Chinese chipmakers more support than they’d otherwise have received.
  • Whoops
• The case of Wuhan Hongxin (HSMC) shows the risk of shoveling money into semiconductors without asking enough questions. According to a Chinese media report that’s since been removed from the internet, HSMC was founded by a group of scam artists who carried fake business cards that read “TSMC—Vice President” and spread rumors that their relatives were top Communist Party officials. They duped the Wuhan local government into investing in their company, then used the funds to hire as CEO TSMC’s former head of R&D. With him on board, they acquired a deep-ultraviolet lithography machine from ASML, then used this feat to raise more funds from investors. But the factory in Wuhan was a shoddily built copy of an old TSMC facility; HSMC was still trying to produce its first chip when the company went bust.
• An official from China’s government planning agency publicly lamented that the country’s chip industry had “no experience, no technology, no talent.” This is an overstatement, but it’s clear that billions of dollars have been wasted in China on semiconductor projects that are either hopelessly unrealistic or, like HSMC, blatant frauds. If China’s Sputnik moment inspires more state-backed semiconductor programs like these, the country won’t be on a path to technological independence.
• Consider, for example, what it would take to replicate one of ASML’s EUV machines, which have taken nearly three decades to develop and commercialize. EUV machines have multiple components that, on their own, constitute epically complex engineering challenges. Replicating just the laser in an EUV system requires perfectly identifying and assembling 457,329 parts. A single defect could cause debilitating delays or reliability problems. No doubt the Chinese government has deployed some of its best spies to study ASML’s production processes. However, even if they’ve already hacked into the relevant systems and downloaded design specs, machinery this complex can’t simply be copied and pasted like a stolen file. Even if a spy were to gain access to specialized information, they’d need a PhD in optics or lasers to understand the science—and even still, they’d lack the three decades of experience accumulated by the engineers who’ve developed EUV.
• EUV machines are just one of many tools that are produced via multinational supply chains. Domesticating every part of the supply chain would be impossibly expensive. The global chip industry spends over $100 billion annually on capital expenditures. China would have to replicate this spending in addition to building a base of expertise and facilities that it currently lacks. Establishing a cutting-edge, all-domestic supply chain would take over a decade and cost well over a trillion dollars in that period.
• However, there’s now a new instruction set architecture called RISC-V that is open-sourced, so it’s available to anyone without a fee. The idea of an open-source architecture appeals to many parts of the chip industry. Anyone who currently must pay Arm for a license would prefer a free alternative. Moreover, the risk of security defects may be lower, because the open nature of an open-source architecture like RISC-V means that more engineers will be able to verify details and identify errors. For the same reason, the pace of innovation may be faster, too. These two factors explain why DARPA has funded a variety of projects related to developing RISC-V. Chinese firms have also embraced RISC-V, because they see it as geopolitically neutral. In 2019, the RISC-V Foundation, which manages the architecture, moved from the U.S. to Switzerland for this reason. Companies like Alibaba are designing processors based on the RISC-V architecture with this in mind.
• Across the chip industry, estimates suggest that China’s share of fabrication will increase from 15 percent at the start of the decade to 24 percent of global capacity by 2030, overtaking Taiwan and South Korea in terms of volume. China will almost certainly still lag technologically. But if more of the chip industry moves to China, the country will have more leverage in demanding technology transfer. It will become more costly for the U.S. and other countries to impose export restrictions, and China will have a broader pool of workers from which to draw. Almost all of China’s chip firms are dependent on government support, so they’re oriented toward national goals as much as commercial ones. “Making profits and going public… are not the priority” at YMTC, one executive told the Nikkei Asia newspaper. Instead, the company’s focused on “building the country’s own chips and realizing the Chinese dream.”
• The Biden administration and most of the media interpreted the chip shortage as a supply chain problem. The White House commissioned a 250-page report on supply chain vulnerabilities that focused on semiconductors. However, the semiconductor shortage wasn’t primarily caused by issues in the chip supply chain. There were some supply disruptions, like COVID lockdowns in Malaysia, which impacted semiconductor packaging operations there. But the world produced more chips in 2021 than ever before—over 1.1 trillion semiconductor devices, according to research firm IC Insights. This was a 13 percent increase compared to 2020. The semiconductor shortage is mostly a story of demand growth rather than supply issues. It’s driven by new PCs, 5G phones, AI-enabled data centers—and, ultimately, our insatiable demand for computing power.
• Besides a massive earthquake—a low but non-zero probability risk—it’s hard to imagine a more severe peacetime shock to supply chains than what the industry has survived since early 2020. The substantial increase in chip production during both 2020 and 2021 is not a sign that multinational supply chains are broken. It’s a sign that they’ve worked.
• The company itself plans to invest over $100 billion between 2022 and 2024 to upgrade its technology and expand chipmaking capacity. Most of this money will be invested in Taiwan, though the company plans to upgrade its facility in Nanjing, China, and to open a new fab in Arizona. Neither of these new fabs will produce the most cutting-edge chips, however, so TSMC’s most advanced technology will remain in Taiwan.
• While Japan could use a new Akio Morita, the United States is in desperate need of a new Andy Grove. America still has an enviable position in the chip industry. Its control over many of the industry’s choke points, including software and machinery, is as strong as ever. Companies like Nvidia look likely to play a foundational role in the future of computing trends like artificial intelligence. Moreover, after a decade in which chip startups were out of fashion, in the past few years Silicon Valley has poured money into fabless firms designing new chips, often focused on new architectures that are optimized for artificial intelligence applications.
• Publicly, Intel is encouraging a new wave of chip nationalism and nervousness about reliance on production in Asia. It’s trying to extract subsidies from both the U.S. and European governments to build fabs at home. “The world needs a more balanced supply chain,” Gelsinger argues. “God decided where the oil reserves are, we get to decide where the fabs are.”
• One recent war game organized by American defense experts envisioned Chinese troops landing on the island and seizing the small Taiwanese garrison there without firing a shot. Taiwan and the U.S. would face the difficult choice of starting a war over an irrelevant atoll or establishing a precedent that China can slice off chunks of Taiwanese territory like pieces of soft salami. “Moderate” responses would include stationing large numbers of U.S. troops in Taiwan or launching cyberattacks on China, both of which could easily escalate into a full-blown conflict.
  • #Anduril
• TSMC’s chairman is certainly right that no one wants to “disrupt” the semiconductor supply chains that crisscross the Taiwan Strait. But both Washington and Beijing would like more control over them. The idea that China would simply destroy TSMC’s fabs out of spite doesn’t make sense, because China would suffer as much as anyone, especially since the U.S. and its friends would still have access to Intel’s and Samsung’s chip fabs. Nor has it ever been realistic that Chinese forces could invade and straightforwardly seize TSMC’s facilities. They’d soon discover that crucial materials and software updates for irreplaceable tools must be acquired from the U.S., Japan, and other countries. Moreover, if China were to invade, it’s unlikely to capture all TSMC employees. If China did, it would only take a handful of angry engineers to sabotage the entire operation. The PLA’s proven it can seize Himalayan peaks from India on the two countries’ disputed border, but grabbing the world’s most complex factories, full of explosive gases, dangerous chemicals, and the world’s most precise machinery—that’s a different matter entirely.
  • #Anduril
• Long gone are the days, as during the 1996 Taiwan Strait crisis, that the U.S. could simply sail an entire aircraft carrier battlegroup through the Strait to force Beijing to stand down. Now such an operation would be fraught with risk for the U.S. warships. Today Chinese missiles threaten not only U.S. ships around Taiwan but also bases as far away as Guam and Japan. The stronger the PLA gets, the less likely the U.S. is to risk war to defend Taiwan. If China were to try a campaign of limited military pressure on Taiwan, it’s more likely than ever that the U.S. might look at the correlation of forces and conclude that pushing back isn’t worth the risk.
  • #Anduril
• If TSMC’s fabs were to slip into the Chelungpu Fault, whose movement caused Taiwan’s last big earthquake in 1999, the reverberations would shake the global economy. It would only take a handful of explosions, deliberate or accidental, to cause comparable damage. Some back-of-the-envelope calculations illustrate what’s at stake. Taiwan produces 11 percent of the world’s memory chips. More important, it fabricates 37 percent of the world’s logic chips. Computers, phones, data centers, and most other electronic devices simply can’t work without them, so if Taiwan’s fabs were knocked offline, we’d produce 37 percent less computing power during the following year.
  • #Anduril
• Yet it would be far easier to find new assembly workers—as difficult as that would be—than to replicate Taiwan’s chipmaking facilities. The challenge wouldn’t simply be building new fabs. Those facilities would need trained personnel, unless somehow many TSMC staff could be exfiltrated from Taiwan. Even still, new fabs must be stocked with machinery, like tools from ASML and Applied Materials. During the 2021–2022 chip shortage, ASML and Applied Materials both announced they were facing delays in producing machinery because they couldn’t acquire enough semiconductors. In case of a Taiwan crisis, they’d face delays in acquiring the chips their machinery requires.
• Taiwan’s president Tsai Ing-wen recently argued in Foreign Affairs that the island’s chip industry is a “ ‘silicon shield’ that allows Taiwan to protect itself and others from aggressive attempts by authoritarian regimes to disrupt global supply chains.” That’s a highly optimistic way of looking at the situation. The island’s chip industry certainly forces the U.S. to take Taiwan’s defense more seriously. However, the concentration of semiconductor production in Taiwan also puts the world economy at risk if the “silicon shield” doesn’t deter China.
  • #Anduril
• There’s no guarantee, of course, that chips will remain as important as they’ve been in the past. Our demand for computing power is unlikely ever to diminish, but we could run out of supply. Gordon Moore’s famous law is only a prediction, not a fact of physics. Industry luminaries from Nvidia CEO Jensen Huang to former Stanford president and Alphabet chairman John Hennessy have declared Moore’s Law dead. At some point, the laws of physics will make it impossible to shrink transistors further. Even before then, it could become too costly to manufacture them. The rate of cost declines has already significantly slowed. The tools needed to make ever-smaller chips are staggeringly expensive, none more so than the EUV lithography machines that cost more than $100 million each.