Palantir Published a Manifesto. Every S&P500 CEO Should Understand Why.
A signaling game, a prisoner's dilemma, and a framework for deciding how much political alignment your business actually needs.
Palantir published a political manifesto that became a bestseller, and most observers read it as corporate political philosophy. This piece reads it as business strategy. Using signaling theory, game theory, and an analysis of corporate behavior under the current administration, it maps the alignment choices American executives are making, why those choices are rational given their specific exposure profiles, and what the calculus looks like when the political cycle turns. The framework applies whether your business depends on federal contracts, faces regulatory intervention, or has no government exposure at all.
The Price of the Table
On April 18, Palantir posted a 22-point summary of its CEO’s recent book-manifesto on X. It has been celebrated in some quarters, while criticism has largely focused on the manifesto’s arguments for Silicon Valley’s moral obligation to arm the military, the coming age of AI deterrence, and the inadequacy of soft power. But those particular arguments cost Palantir Chairman Peter Thiel and CEO Alex Karp absolutely nothing to make. A defense contractor writing that democracies need better weapons is any company encouraging its customers to consume more of its product. The paragraphs that really matter are further down and largely glossed over in coverage, but conveyed an unmistakable signal to those they were designed to impress: that some cultures have produced wonders while others have proven regressive and harmful, that elite intolerance of religious belief exposes the poverty of its political project, and that America has spent half a century avoiding the definition of a national culture in the name of inclusivity. Those kinds of arguments create a permanent public record aligned to a particular political constituency. They attach Palantir and Karp’s name to positions that will be quoted back in the future when the administration changes. These stated positions carry a cost. And that is precisely why they are in the manifesto.
A signal only functions when it is expensive enough that a purely mercenary actor would not send it. Any company seeking federal contracts can publish a white paper on hard power and AI deterrence. What separates a genuine loyalty signal from a marketing brochure is content that carries real personal and reputational cost. The cultural hierarchy argument, the critique of hollow pluralism, the defense of religious belief against secular elite contempt: these are the kinds of passages that make the entire manifesto credible to the current administration, because no business executive optimizing purely for quarterly revenue would think it wise to go there. That credibility is worth something concrete. Palantir’s U.S. government revenue nearly doubled in 2025. The Army’s $10 billion enterprise contract, the DHS billion-dollar omnibus, Maven as a formal program of record: these arrived in the same window as the book. Comparable signals are Zuckerberg reversing DEI policies and absorbing widespread internal dissent, and Bezos blocking his own newspaper’s presidential endorsement and losing 200,000 subscribers. The cost of the signal is the proof of the commitment, and the proof of the commitment is what secures the relationship.
The scale of what is on offer explains the scale of the cost being paid. Palantir's Maven system is embedded in active military targeting. Its software runs across the Army, DHS, ICE, and a growing list of federal agencies. The federal government is not a customer Palantir or its competitors compete for repeatedly in an open market. It is a relationship that, once secured at this depth, becomes structural and self-reinforcing. That is what makes Palantir/Karp’s express ideological commitment rational rather than excessive. A customer worth tens of billions, with switching costs that run into the fabric of national security infrastructure, justifies a level of investment in securing and entrenching the relationship that would look irrational applied to any ordinary commercial account. The manifesto on X, and the book, is that investment. That is the price of an embedded customer relationship that, once secured, does not need to be re-won for a generation.
The Game
What looks like a wave of ideological conversion across American business is better understood as a cascade, and cascades have a specific mechanism. Once one major player makes a costly alignment signal, the calculus shifts for everyone else. The non-signalers are no longer neutral. They are visible as holdouts, and in a political environment where an administration has demonstrated both the willingness and the capacity to target specific companies through tariffs, contract exclusions, and antitrust action, being a visible holdout carries a quantifiable cost. This is not abstract. During President Trump’s first term, the JEDI cloud contract worth $10 billion was awarded to Microsoft after the administration intervened to exclude Amazon, a decision widely attributed to Trump’s hostility toward Bezos and the Washington Post. When the same administration returned to power in 2025, Bezos did not wait to find out whether the lesson still applied.
The underlying structure is a prisoner’s dilemma, and like all prisoner’s dilemmas it produces outcomes that are collectively irrational and individually mandatory. Most CEOs I work with prefer a world in which their business competes on capability alone and political neutrality carries no penalty. But that is not the timeline we live in. Given that other players are signaling, the expected cost of holding out exceeds the expected cost of joining. The dominant strategy becomes alignment even for executives with no genuine ideological sympathy, because the alternative is to be the only person in the room who has not paid a price. This is why the post-election visits to Mar-a-Lago happened fast and in formation, and why the White House’s tech advisory council filled immediately with the CEOs of the most valuable companies in the world.
The administration’s side of this game is a protection structure, and it only works if the threat is occasionally demonstrated. The threat does not need to be applied constantly. It needs to be applied visibly enough that every other player updates their estimate of what non-alignment costs. Intel’s acceptance of a government equity stake was that kind of demonstration. The arrangement came after Trump publicly questioned CEO Lip-Bu Tan’s ties to China and called for his removal. The chip companies that agreed to share a percentage of their China revenues with the federal government in exchange for export permissions were publicly offering a financial settlement. Every Fortune 500 CEO who watched those negotiations understood what they were watching.
Both Thiel and Karp are cerebral leaders who hold degrees in philosophy; Palantir is demonstrating in practice what Michael Spence won the Nobel Prize in economics for formalizing: the value of a signal is a function of its cost, and the cost must be real enough that a party without genuine commitment would not pay it. A million-dollar inauguration donation is not a credible signal of deep alignment because any large company can write that check without thinking twice. A public manifesto arguing that certain cultures are superior to others, that pluralism has become hollow, that the secular elite has lost its legitimacy: that is credible because the reputational cost is asymmetric. It alienates the current administration’s political opponents, which is what makes it visible as a genuine signal. The signal and the cost cannot be separated. The ideology is not decoration on top of a commercial arrangement. It is the very mechanism by which the commercial arrangement is being secured.
Sections one and two were the analysis. What follows is for the senior executive whose business may have government exposure or regulatory risk, and they need to translate the analysis into a decision about their own posture.
The Decision
Every executive watching this should be running the same calculation, and most of them are running it wrong. The mistake is treating alignment as a binary: you either pay or you don’t. The more useful frame has two variables. The first is the degree to which you rely on the federal government for your revenue. The second is your vulnerability to adverse government intervention. Those two variables produce four positions, and each position has a different optimal strategy.
High contract concentration, high vulnerability. This is the most consequential quadrant and the most internally differentiated. Palantir and and a host of IT services contractors like Leidos sit here, but at meaningfully different points. Palantir has depth of deployment, embedded relationships across intelligence and defense agencies, and software running inside active military targeting operations. Those are real switching costs, and they push Palantir toward the Q1 border. Leidos sits deeper in the quadrant: roughly 85 percent of its revenue comes from federal contracts, and its core business in IT services and systems integration is increasingly substitutable as cloud-native alternatives mature and agencies develop in-house capability. The compounding problem for companies in this position is structural. The signal cost rises with each cycle because the administration requires visible proof that the commitment is current, while the underlying competitive position continues to erode. A company here is not buying time and using it. It is buying time while the clock runs faster. The strategic imperative is to build something genuinely hard to replace while the window of political favor is still open, because the relationship alone is worth nothing to the administration that follows.
High contract concentration, low vulnerability. Lockheed Martin, Northrop Grumman, Microsoft, and Anduril share a position that most companies in their orbit would trade for immediately. Their federal contract concentration is high but their vulnerability to government intervention is relatively low, because what they build is either irreplaceable or so deeply embedded in operational infrastructure that displacing it would cost the government more than it would cost them. Lockheed does not need to publish a manifesto. The F-35 program is the manifesto. The alignment signal these companies send is proportionally modest relative to their contract scale, which is itself a form of leverage. When your product is load-bearing, the relationship is more symmetric than it appears from the outside.
Low contract concentration, high regulatory vulnerability. Meta, Google, Apple, TikTok, Amazon and others that have significant B2C exposure each occupy this quadrant for different reasons, but the strategic logic is the same. The threat is not contract cancellation, but that the government will impose significant costs on the business through antitrust action, tariffs, forcing divestiture, or removing legal protections that the business model depends on. Zuckerberg’s alignment tax was an attempt to cap the probability of a structural legal intervention that would cost multiples of whatever the signaling cost him internally. Google faces an active search monopoly ruling. Apple faces App Store antitrust exposure and existential tariff risk on China manufacturing. TikTok faced a divestiture order. The specific threats differ. The calculus is identical: alignment is regulatory insurance, priced against the cost of the intervention it is designed to prevent.
Low contract concentration, low regulatory vulnerability. Shopify, Databricks, HubSpot and many others have no material federal revenue and no significant regulatory overhang. The prisoner’s dilemma pressure is real even here: once the cascade is moving, neutrality begins to look like a statement. But the expected value of paying the alignment tax from this position is genuinely low, and the expected cost is not. A company in this quadrant that sends costly ideological signals is not buying protection from the current administration, yet it is acquiring a liability with a future administration, denominated in relationships, talent, and reputational capital that will need to be rebuilt when the political cycle turns. The discipline required is to recognize that the prisoner’s dilemma only compels alignment when the downside of holding out is concrete and proximate. For companies in this quadrant, it is neither.
Every alignment signal creates a public record, and public records survive administrations. Companies that over-indexed on political relationships without building underlying capability have faced severe reversals when those relationships ended. US firms that built their client base around Washington-backed regimes in Southeast Asia and Latin America during the Cold War found that when those regimes changed, the relationship had no value to whoever came next. The firms with genuinely competitive products found a path forward. The ones whose entire position rested on the political connection did not. The same logic applies here. Deep ideological alignment with a specific administration is a public record that the next administration rarely forgets.
The executives who navigate transitions best are not the ones who hedged most carefully on paper. They are the ones who built something the next administration needs badly enough that continuity is the path of least resistance. Alignment buys time and access. The question is what gets built with that time and access. A company that uses the current window to deepen its operational moat, expand its deployed footprint, and make itself genuinely hard to replace is in a different position when the cycle turns than a company that spent the same window perfecting its loyalty signal. The first company is an asset to whoever runs a future administration. The second is a reminder of who lost.
Adil Husain is a competitive strategist who advises CEOs on how to compete and grow in contested markets. If this piece landed, you can reach him here for a conversation: ahusain@emerging-strategy.com