There’s more: Just after AIPCon, Palantir was added to the S&P 100, an elite group of some of the most closely-watched publicly traded companies. The move was clearly a recognition of Palantir’s ascent: After floundering in the years following its 2020 public market debut, the company’s stock is up 323% from its year-ago price, and its market cap has soared above $200 billion . In short, more than two decades since its founding, Palantir has finally made it big.

Under Karp’s leadership, Palantir has always done things its own way. Founded around the same time as Facebook, YouTube , and Skype, Palantir’s first customer was the U.S. government—primarily defense and intelligence agencies—which meant that its product was shrouded in secrecy from the get-go. (The company’s commercial business currently accounts for 45% of overall revenue, but the other 55% still comes from the government sector.) Still, at its core, there’s nothing mysterious about the company’s tech tools and its army of engineers, which help customers bring digital data from different sources into one unified system. The aim: To “supercharge” all sorts of daily decision-making.

For customers like the Pentagon, that can mean collecting data from space sensors to assist soldiers with real-time warfare strategy. For the commercial sector, the underlying mechanism is pretty much the same, even if the applications are more mundane. One example: Walgreens, another multi-year-contract Palantir customer, is piloting a project that pulls in data from thousands of stores to figure out how and when to reroute pharmacy tasks to other locations, something it calls “workload balancing.” According to Jeff Hoffman, vice president of product at Walgreens, this has led to an increase of about 30% in “operational efficiencies.” That translates to workers being able to utilize that time to “focus on higher-value activities,” Hoffman told the audience at AIPCon—or, in layman’s terms, to make more money.

$2.87 billion

323%

The need for that kind of data analysis is, of course, only growing , and Karp has capitalized on the demand, expanding Palantir’s offerings (one example: its generative AI-powered platform, AIP, which launched in 2023). But he’s also seizing the moment to articulate Palantir’s broader mission to a wider audience—including, critically, its origins as a defense contractor.

In mid-February, Karp and Palantir’s head of corporate affairs, Nicholas Zamiska, published a book titled The Technological Republic . The title offers a critique of what the two call Silicon Valley’s culture of complacency—a fixation on building consumer products like photo-sharing apps and an unwillingness, at least until recently, to partner with or sell to the government. In their New York Times bestseller, the duo asserts that, like Palantir, others in tech have an obligation to protect the institutions that enable the freedoms that have allowed their industry to thrive—the values of Western, liberal democracies. What’s at stake, they argue, is nothing short of our survival.

“Our entire defense establishment and military procurement complex were built to supply soldiers for a type of war—on grand battlefields and with clashes of masses of humans—that may never again be fought,” Karp and Zamiska write. “This next era of conflict will be won or lost with software.” (Conveniently enough, that’s exactly what Palantir makes.)

Why Silicon Valley changed its tune on defense tech

Karp, who has a PhD in neoclassical social theory from Goethe University in Frankfurt, is arguably the most interesting man in software. And when I meet him in his sparse Palo Alto office to talk about his new book, this moment in time for the tech industry, and the growth of Palantir’s business, the conversation ranges in unpredictable directions. Karp references both the Talmud, a lengthy manuscript comprised of Rabbinic debates about the Old Testament, and the New Testament. He also touches on Heine and Goethe, two writers whose works he has read (and memorized large swaths of) in their original German, despite struggling with dyslexia from an early age.

But we keep returning to the themes of beliefs and values—and how malleable those can be even for accomplished leaders. When I ask Karp to make his case for getting those who don’t believe in the values of the government to work with its agencies, he corrects me. “The flaw in the question is you’re making a binary, between people who believe and don’t believe,” he says. “But people are in the ‘tell me what to believe’ category."

Karp asserts that Silicon Valley’s swing towards “wokeness”—characterized by, among other things, the DEI initiatives championed by many companies in 2020 and 2021—didn’t happen because its leaders had firmly-held convictions one way or another. In Karp’s view, those that went all in on it simply realized there were “huge consequences in going against it.”

“It’s not like they [the mainstream in Silicon Valley] are strong on one side or strong on the other, they just want to be left alone,” says Karp. That, he suggests, may explain why many tech companies are now moving away from the DEI efforts they championed just a few years ago. “‘Tell me what the opinion is that will help me be more successful’ is what they really want.”

The case Karp makes in his book and elsewhere, therefore, isn’t about trying to push people or companies into his ideological camp. Rather, it’s about suggesting that they can succeed by emulating Palantir’s strategy—part of which is selling tools that bolster national security efforts, once taboo in Silicon Valley.

“You don’t win your argument by convincing people you’re right, not in Silicon Valley, that’s the wrong axis,” says Karp. “The axis that matters is: Does the business work? Does it perform? What is the quality of the engineers? How valuable will it be in the future? If you win on that axis, you will actually be very convincing.”

Karp himself is not easily swayed by the perpetual shifts in culture and politics. He has been beating the same drum for the last two decades. But, up until recently, very few were marching along with him. “It’s much more that I’ve sharpened and extended my opinions,” he says when I ask if he’s changed his mind on anything “fundamental” to his world view in the last few years.

“You don’t win your argument by convincing people you’re right, not in Silicon Valley. The axis that matters is: Does the business work?"

Indeed, even those who don’t agree with his views would probably say that he’s been consistent, especially when it comes to his willingness to sell defense- and intelligence-related technology to the U.S. government and its allies. Other tech companies, like Alphabet’s Google , have flip-flopped. In 2018, the search giant caved to employee demands and nixed plans to renew a contract with the Pentagon. But more recently, with a rise in “defense tech” companies like drone maker Anduril, an AI arms race that threatens national security, and a political shift among some of tech’s most powerful voices, the tide appears to be turning. Google, for its part, did away with a self-imposed ban on selling AI tools for defense purposes. In February, it explained its rationale, saying it believed that companies, governments, and organizations with common values should work together to “create AI that protects people, promotes global growth, and supports national security.”

“Does everyone now agree with me just because I was saying this 20 years ago?” says Karp. “No, they agree with me because Palantir became a multi-hundred-billion-dollar company and they see the best people in the world want to work here.”

The high-stakes decision to sue the U.S. Army

While Karp’s book has plenty of criticisms for Silicon Valley’s long-time reluctance to engage with the Pentagon, it also takes issue with the government’s slowness to evolve, and to open its procurement processes to new technologies and vendors. Palantir has helped usher in a new era for emerging defense contractors, but it had to break open the door in order to do so. About a decade ago, Palantir sued the U.S. Army, claiming that its contractor selection process had unlawfully shut Palantir out of the competition.

“The most important people in the defense world came in and told me, ‘if you sue the U.S. government, everyone will know you're an asshole,’” says Karp. “To which I responded, ‘I don't feel like I'm an asshole.’”

Karp says that if Palantir had lost its suit, it would have likely been the end for the company. Instead, in late 2016 a federal judge sided with the company, saying that the U.S. Army hadn’t complied with procurement laws.

“It was very high stakes,” Karp says. “But we rose early and killed.”

In 2018, Palantir went on to win a contract with the very entity it had sued, after the Army scrapped its original procurement process and started a new one. (Palantir’s multi-year deal for an “Army Data Platform”, worth upwards of $619 million, was recently extended.) The incident is indicative not just of the company’s relentlessness but also of a key element in Karp’s leadership: Being an asshole—or at least being perceived as one—and being okay with it.

“You know, I run into people all the time, important people, and they're like, ‘I'm really envious of you.’ And I'm like, ‘Well, why?’ Because, honestly, I wouldn't be the first person on my list,” says Karp. “And then, it’s almost invariably, like, ‘Oh man, I wish I just didn't care that people thought I was an asshole.’ And then I’m like, ‘Huh? I didn't realize that’s what you think.’”

Karp says he listens very carefully to why they are saying the things his critics are saying. But, he says, “I don’t have the ability to change what I think because of what they think of me.”

Is defense tech recession-proof?

Karp has obviously benefited from playing the long game. But now that Palantir has made it big, there’s a lot riding on its future—and his ability to continue to show that the business works.

Many analysts have expressed concerns that Palantir’s stock is too expensive, too out of whack with even their most bullish growth estimates.

“Given its strong positioning and execution, there’s no denying that [Palantir] is deserving of a premium valuation,” Gregg Moskowitz, an analyst at investment firm Mizuho, writes in a recent report. “However valuation cannot and should not be irrelevant, and we find it exceedingly difficult to justify [Palantir’s] multiple that in our view already discounts significant further acceleration and upside.” (As of market close on April 14, Palantir traded at about 491 times its trailing 12-month earnings; the broader Nasdaq 100, in contrast, traded at about 33 times earnings.)

It’s also not clear what the impact of the second coming of the Trump administration will be on companies that count on defense as a significant revenue stream. Already, President Trump has instructed the Pentagon to trim expenses . What’s more, the recent tariffs-induced market volatility has wreaked havoc on pretty much all industries—and a recession would certainly impact Palantir’s business. Even in Europe, which is about to undertake its greatest rearming effort since World War II, and where the company already sells its software to several governments, the growth prospects for Palantir aren’t necessarily rosy.

"Nearly everyone has decided that it would suicidal not to support the U.S. government and its defense efforts."

“The most important thing in tech is not how much you spend, but with whom,” says Karp. “It's really important for Europe to end up with a world class defense industry. So how do you build that? How do you relearn those skills? Modern fighting is software, hardware, hybrid systems.”

Back in the U.S., Karp says software spend in defense budgets is still a tiny fraction of what it should be (Palantir estimates the figure stands at under 1%). But he believes we now have the very best and most innovative companies here, and increasingly ready to partner with the government.

“I think the tech industry is now positive towards working with the U.S. government,” says Karp. “Does everyone truly believe? I don’t know, I can’t look into their hearts. But nearly everyone has decided that it would suicidal not to support the U.S. government and its defense efforts.”

For Palantir, this has never been in question—even as its commercial business expands, the company’s partnership with the government remains a critical part of both its revenue stream and its story, and of course of Karp’s own mission and narrative. And whether you agree with him or not, you have to admit: Karp has been putting on the same show for decades now, whether in front of naysayers or fanboys.

This story was originally featured on Fortune.com