The rise of the American software industry in the 20th century was made possible by a partnership between emerging technology companies and the U.S. government. Silicon Valley’s earliest innovations were driven not by technical minds chasing trivial consumer products but by scientists and engineers who aspired to address challenges of industrial and national significance using the most powerful technology of the age. Their pursuit of breakthroughs was intended not to satisfy the passing needs of the moment but rather to drive forward a much grander project, channeling the collective purpose and ambition of a nation.

This early dependence of Silicon Valley on the nation-­state and indeed the U.S. military has, for the most part, been forgotten, written out of the region’s history as an inconvenient and dissonant fact—­one that clashes with the Valley’s conception of itself as indebted only to its capacity to innovate. The United States since its founding has always been a technological republic, one whose place in the world has been made possible and advanced by its capacity for innovation.

But there is also another essential element of American success. It was a culture, one that cohered around a shared objective, that won the last world war. And it will be a culture that wins, or prevents, the next one.

This essay has been excerpted from Karp and Zamiska’s new book, The Technological Republic.

At present, however, the principal shared features of American society are not civic or political but rather cohere around entertainment, sports, celebrity, and fashion. This is not the result of some unbridgeable political division. The interpersonal tether that makes possible a form of imagined intimacy among strangers within groups of a significant size was severed and banished from the public sphere. The old means of manufacturing a nation—the civic rituals of an educational system, mandatory service in national defense, religion, a common language, and a free and thriving press—have all but been dismantled or withered from neglect and abuse. This distaste for collective experience and endeavor made America, and American culture, vulnerable.

The establishment left has failed its cause and thoroughly eroded its potential. The frenetic pursuit of a shallow egalitarianism in the end hollowed out its broader and more compelling political project. What we need is more cultural specificity ­in education, technology, and politics—­not less. The vacant neutrality of the current moment risks allowing our instinct for discernment to atrophy. Only the resurrection of a shared culture, not its abandonment, will make possible our continued survival and cohesion. And only by combining the pursuit of innovation with the shared objectives of the nation can we both advance our welfare and safeguard the legitimacy of the democratic project itself.

Silicon Valley once stood at the center of American military production and national security. Fairchild Camera and Instrument Corporation, whose semiconductor division was founded in Mountain View, California, and made possible the first primitive personal computers, built reconnaissance equipment for spy satellites used by the CIA beginning in the late 1950s. For a time after World War II, all of the U.S. Navy’s ballistic missiles were produced in Santa Clara County, California. Companies such as Lockheed Missiles and Space, Westinghouse, Ford Aerospace, and United Technologies had thousands of employees working in Silicon Valley on weapons production through the 1980s and into the 1990s.

This union of science and the state in the middle part of the 20th century began in earnest during World War II. In November 1944, as Soviet forces closed in on Germany from the east, President Franklin D. Roo­se­velt was in Washington, D.C., already contemplating an American victory and the end of the conflict that had remade the world. Roo­sevelt sent a letter to Vannevar Bush, a pastor’s son who had become the head of the U.S. Office of Scientific Research and Development, where he helped lead the Manhattan Project.

In the letter, Roo­se­velt described “the unique experiment” that the United States had undertaken during the war to leverage science in service of military ends. Roo­sevelt anticipated the next era—­and partnership between national government and private industry—­with precision. He wrote that there was “no reason why the lessons to be found in this experiment”—­that is, directing the resources of an emerging scientific establishment to help wage the most significant and violent war that the world had ever known—­“cannot be profitably employed in times of peace.”

Roosevelt’s ambition was clear. He intended to see the machinery of the state—its power and prestige, as well as the financial resources of the newly victorious nation and emerging hegemon—spur the scientific community forward in service of, among other things, the advancement of public health and national welfare. The challenge was to ensure that the engineers and researchers who had directed their attention to the industry of war—and particularly the physicists, who, as Bush noted, had “been thrown most violently off stride”—­could shift their efforts back to civilian advances in an era of relative peace.

The entanglement of the state and scientific research both before and after the war was itself built on an even longer history of connection between innovation and politics. Many of the earliest leaders of the American republic were inventors, including Thomas Jefferson, who designed sundials and studied writing machines, and Benjamin Franklin, who experimented with and constructed objects as varied as lightning rods and eyeglasses.

Unlike the legions of lawyers who have come to dominate American politics in the modern era, many early American leaders, even if not practitioners of science themselves, were nonetheless remarkably fluent in matters of engineering and technology. John Adams, the second president of the United States, was, by one historian’s account, focused on steering the early republic away from “unprofitable science, identifiable in its focus on objects of vain curiosity,” and toward more practical forms of inquiry, including “applying science to the promotion of agriculture.”

Many of the innovators of the 18th and 19th centuries were polymaths whose interests diverged wildly from the contemporary expectation that depth, as opposed to breadth, is the most effective means of contributing to a field. The frontiers and edges of science were still in that earliest stage of expansion that made possible and encouraged an interdisciplinary approach, one that would be almost certain to stall an academic career today. That cross-­pollination, as well as the absence of a rigid adherence to the boundaries between disciplines, was vital to a willingness to experiment, and to the confidence of political leaders to opine on engineering and technical questions that implicated matters of government.

The rise of J. Robert Oppenheimer and dozens of his colleagues in the late 1930s further situated scientists and engineers at the heart of American life and the defense of the democratic experiment. Joseph Licklider, a psychologist whose work at the Massachusetts Institute of Technology anticipated the rise of early forms of AI, was hired in 1962 by the organization that would become the U.S. Defense Advanced Research Projects Agency—­an institution whose innovations would include the precursors to the modern internet as well as the global positioning system. His research for his now classic paper “Man-­Computer Symbiosis,” which was published in March 1960 and sketched a vision of the interplay between computing intelligence and our own, was supported by the U.S. Air Force.

There was a closeness, and significant degree of trust, in the relationships between political leaders and the scientists on whom they relied for guidance and direction. Shortly after the launch by the Soviet Union of the satellite Sputnik in October 1957, Hans Bethe, the German-­born theoretical physicist and adviser to President Dwight Eisenhower, was called to the White House. Within an hour, there was agreement on a path forward to reinvigorate the American space program. “You see that this is done,” Eisenhower told an aide. The pace of change and action in that era was swift. NASA was founded the following year.

By the end of World War II, the blending of science and public life—­of technical innovation and affairs of state—­was essentially complete and unremarkable. Many of these engineers and innovators would labor in obscurity. Others, however, were celebrities in a way that might be difficult to imagine today. In 1942, as war spread across Europe and the Pacific, an article in Collier’s introduced Vannevar Bush, who was at the time a little-­known engineer and government bureaucrat, to the magazine’s readership of nearly 3 million, describing Bush as “the man who may win the war.” (Three years later, Bush published “As We May Think” in The Atlantic, praising scientists for working together in a “common cause,” and anticipating many aspects of the information age that lay ahead.) Albert Einstein was not only one of the 20th century’s greatest scientific minds but also one of its most prominent celebrities—a popular figure whose image and breakthrough discoveries, which so thoroughly defied our intuitive understanding of the nature of space and time, routinely made front-­page news. And it was often the science itself that was the focus of coverage.

[From the February 1949 issue: J. Robert Oppenheimer’s ‘The Open Mind’]

This was the American century, and engineers were at the heart of the era’s ascendant mythology. The pursuit of public interest through science and engineering was considered a natural extension of the national project, which entailed both protecting U.S. interests and moving society—indeed, civilization—up the hill. And while the scientific community required funding and extensive support from the government, the modern state was equally reliant on the advances that those investments in science and engineering produced. The technical outperformance of the United States in the 20th century—­that is, the country’s ability to reliably deliver economic and scientific advances for the public, whether medical breakthroughs or military capabilities—­was essential to its credibility.

As the philosopher Jürgen Habermas has suggested, a failure by leaders to deliver on implied or explicit promises to the public has the potential to provoke a crisis of legitimacy for a government. When emerging technologies that give rise to wealth do not advance the broader public interest, trouble often follows. Put differently, the decadence of a culture or civilization, and indeed its ruling class, will be forgiven only if that culture is capable of delivering economic growth and security for the public. In this way, the willingness of the engineering and scientific communities to come to the aid of the nation has been vital not only to the legitimacy of the private sector but to the durability of political institutions across the West.

The modern incarnation of Silicon Valley has strayed significantly from this tradition of collaboration with the U.S. government, focusing instead on the consumer market, including the online advertising and social-media platforms that have come to dominate—­and limit—our sense of the potential of technology. A generation of founders cloaked themselves in the rhetoric of lofty and ambitious purpose—their rallying cry that they intend “to change the world” has grown lifeless from overuse—­but many of them raised enormous amounts of capital and hired legions of talented engineers merely to build photo-­sharing apps and chat interfaces for the modern consumer.

A skepticism of government work and national ambition took hold in the Valley. The grand, collectivist experiments of the middle of the 20th century were discarded in favor of a narrow attentiveness to the desires and needs of the individual. The market rewarded shallow engagement with the potential of technology, as start-up after start-up catered to the whims of late-capitalist culture without any interest in constructing the technical infrastructure that would address our most significant challenges as a nation. The age of social-media platforms and food-delivery apps had arrived. Medical breakthroughs, education reform, and military advances would have to wait.

[Read: The divide between Silicon Valley and Washington is a national-security threat]

For decades, the U.S. government was viewed in Silicon Valley as an impediment to innovation and a magnet for controversy—more an obstacle to progress than its logical partner. The technology giants of the current era long avoided government work. The level of internal dysfunction within many state and federal agencies created seemingly insurmountable barriers to entry for outsiders, including the insurgent start-ups of the new economy. In time, the tech industry lost interest in politics and broader collaborations. It viewed the American national project, if it could even be called that, with a mix of skepticism and indifference. As a result, many of the Valley’s best minds, and their flocks of engineering disciples, turned to the consumer for sustenance.

The interests and political instincts of the American elite diverged from those of the rest of the country following the end of World War II. The economic struggles of the country and geopolitical threats of the 20th century today feel distant to most software engineers. The most capable generation of coders has never experienced a war or genuine social upheaval. Why court controversy with your friends or risk their disapproval by working for the U.S. military when you can retreat into the perceived safety of building another app?

As Silicon Valley turned inward and ­toward the consumer, the U.S. government and the governments of many of its allies scaled back involvement and innovation across numerous domains, including space travel, military software, and medical research. The state’s retreat left a widening innovation gap. Many cheered this divergence: Skeptics of the private sector argued that it could not be trusted to operate in public domains while those in the Valley remained wary of government control and the misuse or abuse of their inventions. For the United States and its allies in Europe and around the world to remain as dominant in this century as they were in the previous one, however, they will require a union of the state and the software industry—­not their separation and disentanglement.

[Read: The crumbling foundation of America’s military]

Indeed, the legitimacy of the American government and democratic regimes around the world will require an increase in economic and technical output that can be achieved only through the more efficient adoption of technology and software. The public will forgive many failures and sins of the political class. But the electorate will not overlook a systemic inability to harness technology for the purpose of effectively delivering the goods and services that are essential to our lives.

In late 1906, Francis Galton, a British anthropologist, traveled to Plymouth, En­gland, in the country’s southwest, where he attended a livestock fair. His interest was not in purchasing the poultry or cattle that were available for sale at the market but in studying the ability of large groups of individuals to correctly make estimates. Nearly 800 visitors at the market had written down estimates of the weight of a particular ox that was for sale. Each person had to pay six pennies for a chance to submit their guess and win a prize, which deterred, in Galton’s words, “practical joking” that might muddy the results of the experiment. The median estimate of the 787 guesses that Galton received was 1,207 pounds, which turned out to be within 0.8 percent of the correct answer of 1,198 pounds. It was a striking result that would prompt more than a century of research and debate about the wisdom of crowds and their ability to more accurately make estimates, and predictions, than a chosen few. For Galton, the experiment pointed to “the trustworthiness of a democratic judgment.”

But why must we always defer to the wisdom of the crowd when it comes to allocating scarce capital in a market economy? We seem to have unintentionally deprived ourselves of the opportunity to engage in a critical discussion about the businesses and endeavors that ought to exist, not merely the ventures that could. The wisdom of the crowd at the height of the rise of Zynga and Groupon in 2011 made its verdict clear: These were winners that merited further investment. Tens of billions of dollars were wagered on their continued ascent. But there was no forum or platform or meaningful opportunity for anyone to question whether our society’s scarce resources ought to be diverted to the construction of online games or a more effective aggregator of coupons and discounts. The market had spoken, so it must be so.

Americans have, as Michael Sandel of Harvard has argued, been so eager “to banish notions of the good life from public discourse,” to require that “citizens leave their moral and spiritual convictions behind when they enter the public square,” that the resulting void has been filled in large part by the logic of the market—­what Sandel has described as “market triumphalism.” And the leaders of Silicon Valley have for the most part been content to submit to this wisdom of the market, allowing its logic and values to supplant their own. It is our own temerity and unwillingness to risk the scorn of the crowd that have deprived us of the opportunity to discuss in any meaningful way what the world we inhabit should be and what companies should exist. The prevailing agnosticism of the modern era, the reluctance to advance a substantive view about cultural value, or lack thereof, for fear of alienating anyone, has paved the way for the market to fill the gap.

The drift of the technological world to the concerns of the consumer both reflected and helped reinforce a certain technological escapism—­the instinct by Silicon Valley to steer away from the most important problems we face as a society and toward what are essentially the minor and trivial yet solvable inconveniences of everyday consumer life: such as online shopping and food delivery. An entire swath of arenas, including national defense, violent crime, education reform, and medical research, appeared too intractable, too thorny, and too politically fraught to address in any real way. (This was the challenge we have aimed to address at Palantir—to build technology that serves our mos significant and vital needs, including those of U.S. defense and intelligence agencies, instead of merely catering to the consumer.)

Most were content to set the hard problems aside. Consumer apps and trinkets did not talk back, hold press conferences, or fund pressure groups. The tragedy is that serving the consumer rather than the public has often been far easier and more lucrative for Silicon Valley, and certainly less risky.

The path forward will involve a reconciliation of a commitment to the free market, and its atomization and isolation of individual wants and needs, with the insatiable human desire for some form of collective experience and endeavor. Silicon Valley offered a version of this combination. The Sunnyvales, Palo Altos, and Mountain Views of the world were company towns and city-­states, walled off from society and offering something that the national project could no longer provide. Technology companies formed internally coherent communities whose corporate campuses attempted to provide for all of the wants and needs of daily life. They were at their core collectivist endeavors, populated by intensely individualistic and freethinking minds, and built around a set of ideals that many young people craved: freedom to build, ownership of their success, and a commitment above all to results.

Other nations, including many of our geopolitical adversaries, understand the power of affirming shared cultural traditions, mythologies, and values in organizing the efforts of a people. They are far less shy than we are about acknowledging the human need for communal experience. The cultivation of an overly muscular and unthoughtful nationalism has risks. But the rejection of any form of life in common does as well. The reconstruction of a technological republic, in the United States and elsewhere, will require a re-embrace of collective experience, of shared purpose and identity, of civic rituals that are capable of binding us together. The technologies we are building, including the novel forms of AI that may challenge our present monopoly on creative control in this world, are themselves the product of a culture whose maintenance and development we now, more than ever, cannot afford to abandon. It might have been just and necessary to dismantle the old order. We should now build something together in its place.

This essay has been excerpted from Alexander C. Karp and Nicholas W. Zamiska’s new book, The Technological Republic.

On Tuesday morning, I was at the United States Military Academy, in West Point, for a national-security conference. I was invited to observe several classes of cadets in a comparative-politics course. At the end of each class, their instructor, a young Special Forces major who had already seen a variety of conflicts during his short career, would give the students the chance to ask me questions about the “real world” Army. I find such interactions with cadets to be fun and engaging. After warming up, they always seem to get around to asking: What are the hot spots in the world, and where do you see us serving in support of our country during our career?

In answering, I give them my own history. Entering West Point in 1971, my class expected to serve in Vietnam. But by the time we’d graduated, five years later, America was out of that war. Most of us were instead sent to Europe, preparing for a clash between the Soviets and NATO that never arrived. After the Berlin Wall came down, we thought our nation would be at peace for years—but then the Army, which had trained for years to defend the border of Germany, found itself attacking Iraqi forces in Desert Storm. Then, after 9/11, we conducted counterterror and counterinsurgency fights in two distinctly different countries we’d never expected to go to.

The message for the cadets? As soldiers, prepare yourself for anything. Go where you’re sent, conduct operations to the best of your ability, serve your nation well, and follow your oath to defend the Constitution.

That night, in my hotel room, I watched the president stand next to the Israeli prime minister and suggest where the next generation of U.S. soldiers might go. Most Americans were surprised by Donald Trump’s announcement that the United States would be “taking over Gaza”—that we would clear unexploded ordnance, “level the site,” deploy U.S. troops if necessary, and turn the area into the “Riviera of the Middle East.”

After the press conference, I watched analysts, human-rights activists, and Middle Eastern officials condemn the proposed displacement of Palestinians, with some seeing in the proposal a violation of international law.

As stunned as I was by Trump’s announcement, my first thought was: If the military were told to deploy, how in the heck would it do this mission? As a commander, I had been assigned some tough missions. And I remembered reading that when General George C. Marshall made Dwight D. Eisenhower the commander of the European invasion force in World War II, he gave him the succinct written order to “enter the continent of Europe and defeat the Nazi war machine.” Eisenhower wrote that he was immediately overwhelmed by the scope and scale of that mission, the resources that it would require, and the operational environment—enemy, allies, terrain—the troops would face. But he started his planning, and eventually executed the D-Day landings. If the U.S. could pull that off, how much more difficult would it be for it to take over Gaza and turn it into the Riviera of the Middle East?

Answering that question requires translating Trump’s general directive into the specifics of implementation. First, consider the dimensions of the Gaza Strip. At approximately 141 square miles, it’s six times the size of Manhattan. Just razing half-destroyed structures, ridding the area of rubble, and erecting new buildings would present a daunting engineering challenge. One engineer I spoke with that night estimated that the cost would be $30 billion to $80 billion. That broadly aligns with the United Nations Development Program’s estimate of $30 billion to $40 billion.

The UN also estimates that the war has left more than 50 million tons of rubble in Gaza, and claims that clearing that debris could take more than 15 years. Given that more than 170,000 buildings have been damaged or destroyed during the Israeli military’s months-long operations, replacing them would require thousands of civil, structural, electrical, and environmental engineers coordinating their various disciplines, and could take decades.

Who would do this work? One cable pundit suggested using the United States Army Corps of Engineers. That organization has only about 35,000 civilian and 700 military personnel, and is already stretched thin by its domestic responsibilities. Deploying even a portion of the Army Corps would require a considerable commitment, and extensive support for the force within the area of its operation. Such an endeavor would likely necessitate extensive coordination with international partners and private organizations to manage the substantive building effort.

What the military calls the operational environment, or OE, is something that any commander or manager must assess before committing forces to an area. The Gaza/Israel OE is fraught with dangers that would affect and impede the already-challenging rubble-removal and construction efforts. The intelligence needed to counter the activity of terrorist groups comes from a variety of agencies, including the CIA, the FBI, and the Defense Intelligence Agency. All of these are undergoing personnel scrubs by DOGE, which is interrupting the flow of information that would help identify, prevent, and respond to threats. Dismissing senior counterterrorism leaders at the FBI, or encouraging large numbers of CIA officers to leave the agency, will weaken international intelligence sharing, risk increasing terror activity, and heighten international-security risks.

Almost-uniform international opposition would further complicate the challenge. Palestinian leaders were quick to denounce the demand that more than 2 million Gazans leave their home. Our European allies—on whom we depend for support in the area—met Trump’s comments with skepticism and criticism. UN Secretary-General António Guterres said that forced displacement would be “tantamount to ethnic cleansing.” And the proposal also aroused domestic opposition, including from some of Trump’s closest political allies.

When Eisenhower was told to storm the European continent, he could count on the support of the American people, our allies, and our intelligence services—and could draw on enormous resources made available for the task. If the Gaza mission were to be handed to the military today, it would enjoy none of those advantages.

I purposely have not provided what the military calls a troop-to-task requirement, or an analysis of the number of soldiers that would be needed to accomplish the mission. And if troops did go to Gaza as part of this operation, as the president said might be needed, some of them would likely never return—just as thousands didn’t return from storming the beaches during World War II. Before the current administration goes any further, it should take stock of that reality.

As Elon Musk set out to upend the federal bureaucracy on behalf of Donald Trump, he reached out to Trump’s team with an unusual request: U.S. law generally prohibits noncitizens from working for the federal government, but Musk was hoping to make an exception for Baris Akis, a Turkish-born venture capitalist with a green card who had become a close ally.

The answer, delivered privately by Trump’s advisers, was an unequivocal no, two people familiar with the decision told us.

Trump’s White House is in the business of deporting people, and bringing in a foreign national to help shrink the government’s American workforce would send a confusing message, one of these people said. (Neither Akis nor the White House responded to a request for comment.)

Musk and his team accepted the rejection and moved on, but the previously unreported exchange offers a glimpse into the complex dynamics of the Musk-Trump relationship, arguably the most consequential partnership in Washington. This story is based on interviews with six people who have worked closely with Trump or Musk or are directly familiar with their relationship, all of whom requested anonymity to describe private interactions.

The world’s richest man has established himself as a singular force in the administration’s effort to slash government programs, agencies, and federal employees. Yet as an unelected “special government employee,” Musk still relies on the president for his authority. Since sweeping into Washington alongside Trump, Musk has wielded enormous power. He has pressured federal employees into deferred resignations; dug into government data and financial systems; used his massive social-media platform, X, to pick fights and bully opponents; and fed the U.S. Agency for International Development “into the woodchipper,” as he boasted yesterday on X, or at least helped get the agency folded into the State Department. He believes that understanding and mastering the government’s computer systems is the key to overhauling and fixing the government. But he does so at Trump’s behest, at least for as long as he has the president’s blessing.

Trump has made a point in recent days of making clear his supremacy over Musk, and Musk, for all his influence, has found himself bending to the strictures of the White House. Musk’s private security team, for instance, must wait in the parking lot at the 1600 Pennsylvania Avenue campus when he goes to work in a conference room at the Eisenhower Executive Office Building because of the building’s own security protocols, one of the people familiar with the arrangement told us.  

“Elon can’t do and won’t do anything without our approval. And we will give him the approval where appropriate, and where not appropriate we won’t,” Trump said yesterday during a signing event in the Oval Office. “Where we think there is a conflict or there’s a problem, we won’t let him go near it.”

On Sunday, Trump exited Air Force One after arriving in Washington from Palm Beach, and praised Musk as a “cost-cutter” who was “doing a good job,” before establishing the hierarchy: “Sometimes we won’t agree with it, and we’ll not go where he wants to go.”

Trump’s somewhat pointed comments on Musk “are important,” a longtime Trump confidant told us, explaining that “there’s one president.” This person said that Trump had learned about how to work within the government during his first term, but “that’s not true of Elon.”

But Musk nonetheless has threatened to steal the spotlight from Trump in recent days, becoming the public face of the administration’s most disruptive moves, including an effort to force thousands of voluntary resignations from the federal workforce. Inside the West Wing, he has found many allies, having ingratiated himself with mid-tier Trump aides early in the transition, when he moved down to Trump’s private Mar-a-Lago club and established himself as an accessible, if quirky, presence. He regularly shared his cellphone number, including with younger staff, and spent his days sending around memes and ideas about overhauling the government, according to a person who saw the texts but who was not authorized to speak publicly.

Musk has a group of loyalists he often brings with him to each of his various government projects—a cadre Trump today praised as “smart people.” But, unlike many people with his net worth and renown, Musk “travels pretty light,” one person told us. Two people told us that, during the time he spent at Mar-a-Lago, they most regularly saw him with his young son X—“just that kid on his shoulders,” one of the people said—and sometimes X’s nanny.

Musk preempted Trump shortly after midnight Monday in a live broadcast on X with the announcement that the Trump administration would seek to shutter USAID based on his own team’s investigation of the agency. “As we dug into USAID, it became apparent that what we have here is not an apple with a worm in it,” Musk said. “What we have here is just a ball of worms.”

And he has continued to pick public fights with Democratic leaders despite his new day job as a government employee. He accused House Democratic leader Hakeem Jeffries of wanting to continue “waste fraud and abuse” after Jeffries attacked Musk’s leadership of Trump’s Department of Government Efficiency, which operates as a part of the Executive Office of the President. Musk also wrote on X that Senate Democratic leader Chuck Schumer was mad about his work “dismantling the radical-left shadow government.”

Musk’s decision to seek a special dispensation for a noncitizen adviser came after the Tesla CEO’s views on immigration policy became a flash point inside the president’s circle late last year. When Musk defended the practice of giving H-1B visas to highly skilled foreign workers, Trump allies including Steve Bannon attacked Musk as part of a group of “techno-feudalists” undermining American workers. Trump had previously been critical of the H-1B visa system, but eventually sided with Musk in the dispute. Musk also moderated his stance, calling for “major reform” to how the visas are granted.

Democrats, who have struggled to respond effectively to Trump in the first weeks of his second term, have become more focused on Musk as a potential weak point for the president, as polling has shown significant public concern about Musk. A late-January poll by Quinnipiac University found that 53 percent of voters disapproved of him playing a prominent role in the Trump administration, compared with 39 percent who approved. About one in five Republican voters disapproved of Musk’s role.

Because of his special-government-employee status, Musk’s time in government is expected to be limited. Employees under this status, who do not have to divest from outside conflicts of interest, are permitted to work no more than 130 days in a single year. Other members of Musk’s team, including Katie Miller, a Department of Government Efficiency adviser, are also working for the government under the temporary designation. (Musk is also close to Miller’s husband, White House deputy chief of staff Stephen Miller.)

Musk’s preferred work habits before entering the federal bureaucracy could provide a window into how he might continue to feed “into the woodchipper” programs and spending that he views as Washington bloat. During his time in the private sector, Musk tended to burrow into each of his companies on different days, a person familiar with his routine told us. Monday was for SpaceX, Tuesday was for Tesla, and Wednesday was for X.

This week, Sunday and Monday were clearly for USAID. By tomorrow and Thursday, however, he might be ready to turn his buzz saw elsewhere.

Sometimes a constitutional crisis sneaks up on you, shrouded in darkness, revealing itself gradually. Other times it announces itself dramatically. Elon Musk, to whom Donald Trump has delegated the task of neutering the congressional spending authority laid out in Article I of the Constitution, could hardly be more obvious about his intentions if he rode into Washington on a horse trailed by Roman legions.

“This is the one shot the American people have to defeat BUREAUcracy, rule of the bureaucrats, and restore DEMOcracy, rule of the people,” Musk wrote at 3:59 a.m. today on his social-media platform. “We’re never going to get another chance like this. It’s now or never. Your support is crucial to the success of the revolution of the people.” Here is Musk, as proxy for Trump, casting himself as a revolutionary force and embodiment of the popular will, demanding extraordinary powers to fight some unstated emergency.

Why, exactly, is eliminating these programs right this very instant so important? If, as Musk says, they are teeming with waste and fraud, presumably Congress could pass legislation to reduce or eliminate the problem, and if that were to fall short, it could try again later. Instead, Musk cites a vague crisis that requires suspending normal operations and concentrating power in his own hands. According to various reports, he is holed up in the Eisenhower Building with a small team of young engineers who possess neither government experience nor the authority to question his impulsive judgments, on the hunt for Marxist plots lurking within long-standing federal programs.

[Jonathan Lemire: Elon Musk is president]

The situation exposes a well-known flaw in the design of the Constitution. The Founders, famously, failed to anticipate the rise of political parties. They assumed that each branch of government would jealously guard its own powers, and thus check the others. But political parties created a different incentive system, in which members of the legislative branch can see their role as essentially employees of the president. Trump, who has convinced the Republican base that his interests are indistinguishable from the party’s and transposed his overbearing Apprentice boss persona onto his relations with co-partisans in Congress, is exploiting these incentives more than any other president in history.  

In theory, Congress ought to revolt against the prospect of Musk deciding which federal programs should live and which should die. In reality, its members largely share Trump’s goals—and to the extent that they don’t, they correctly fear that opposing him would invite a primary challenge. What’s more, this particular constitutional crisis has an inherent partisan asymmetry. If Trump and Musk succeed in taking the power of the purse from Congress, they will effectively reset the rules of the game in favor of the right. Congress’s spending powers would be redefined as setting a ceiling on spending, but not a floor. A world in which the president could cut spending without exposing Congress to accountability would hand small-government conservatives the opportunity to carry out policies they’ve long desired but been too afraid to vote for.

And so, although a handful of conservative intellectuals, including the budget wonk Brian Riedl of the Manhattan Institute and the law professor and former Bush-administration lawyer Jack Goldsmith, have described Musk’s ambitions as unconstitutional, most of the establishment right has cheered him on or stayed quiet. Senator Thom Tillis of North Carolina conceded that Musk’s project might not be strictly constitutional, but nonetheless told the news site NOTUS that “nobody should bellyache about that.”

Making things even more disturbing is the chaotic legal gray area in which Musk is operating. Musk and his team are working in secret, without hearings or public debate. According to Wired, they gained access to the Treasury Department’s federal payment system, shoving aside the long-time staffer overseeing it and ignoring its safety protocols. Democrats suspect that Musk is breaching numerous federal laws, but without any oversight, it is hard to tell precisely what he is doing. In any case, Musk might not have much reason to care about following the law. Trump has already made plain, by issuing mass pardons and commutations for the January 6 insurrectionists, that he will protect illegal conduct on his behalf.

Meanwhile, Musk has adopted Trump’s habit of deeming opposition to his actions inherently criminal. He has called the United States Agency for International Development, a decades-old program with support in both parties, a “criminal organization.” After an X user posted the names of the young engineers working with Musk, previously reported by Wired, he responded, “You have committed a crime.” The X user’s account has since been suspended.

Reporting on the identities of powerful public officials is, in fact, not a crime—even, or especially, if those officials have assumed public powers without going through formal channels. Musk has nonetheless gotten backup for his threats from Edward R. Martin Jr., a former “Stop the Steal” organizer whom Trump installed as U.S. attorney for the District of Columbia. In a vague but menacing message posted (naturally) on X, Martin warned that “certain individuals and/or groups have committed acts that appear to violate the law in targeting DOGE employees.” Martin declined to identify either the individuals or the laws they’d allegedly broken, nor did he acknowledge that reporting about or criticizing Musk’s work constitutes First Amendment–protected activity. Whether Martin acts upon these threats remains to be seen. In the meantime, however, he is contributing to the atmosphere of menace surrounding Trump and Musk by delivering their threats with a legal sheen, like some kind of MAGA Tom Hagen.

[Read: The ‘rapid unscheduled disassembly’ of the United States government]

The courts will have the final say over Trump’s audacious power grab. In all likelihood, they will affirm congressional authority to set spending levels authorized by the Constitution. But the Constitution ultimately means whatever five Supreme Court justices say it means. The Court’s more conservative justices often apply the most right-wing interpretation of the text they can plausibly defend, and occasionally one they can’t plausibly defend.

What’s more, Musk seems to have intuited that he can destroy programs and bureaucratic cultures faster than the system can restore them. Firing officials en masse, throwing the people and clients that rely on those programs into confusion and financial risk, and striking fear into the whole federal apparatus can break down the institutions and destroy their institutional knowledge. Rebuilding is painfully slow; destruction is rapid. This may be the dynamic Musk has in mind when he insists that his work must happen “now or never.”

Not even the most committed small-government-conservative lawmaker would design a process like the one now occurring: a handful of political novices, many of them drinking deep from the fetid waters of right-wing conspiracy theorizing, tearing through the federal budget, making haphazard decisions about what to scrap. And indeed, no elected body has designed this process. Trump and Musk have arrogated the power to themselves. The true urgent cause is to return that power to the legislature before the damage becomes irreversible.