In 2000, Yahoo found itself the target of French ire. In the United States, someone was using Yahoo to sell Nazi memorabilia — an act illegal in France — prompting a French court to demand the search engine’s removal of the seller and items. Yahoo scoffed, rejecting the idea that it was beholden to another nation’s laws. “Asking us to filter access to our sites according to the nationality of web surfers,” explained founder Jerry Yang, is “very naïve.”1 But all it took was the threat of considerable fines for Yahoo to change its tune and pull the materials.
More significant concessions to foreign powers were to come. In 2002, Yahoo agreed to the Chinese government’s demand that it block sensitive information, and in 2005 it provided the CCP with the identity of a dissident journalist who had used Yahoo Mail, leading to the reporter’s imprisonment. Yang explained, “To be doing business in China, or anywhere else in the world, we have to comply with local law. … [W]e have to follow the law.”2
In only a few years, a leading tech company transformed from sovereign pioneer to prostrate lackey for any country demanding compliance. The lesson for the tech industry should have been clear: you may not be interested in politics, but politics is interested in you. Hopes of transcending borders and escaping to digital freedom will dissipate against old-fashioned political authority and the threat of force.
Recent years have seen a new, and uniquely audacious, iteration of this mistaken hope: venture capitalist Balaji Srinivasan’s “network state” — a “highly aligned online community with a capacity for collective action that crowdfunds territory around the world and eventually gains diplomatic recognition from pre-existing states.”3 Originating in a lecture at Y Combinator in 2013, expanded into a self-published book in 2022, and now gaining momentum through a series of conferences, the network state offers a bold thesis about the relationship between technology and governance.4 Network statists dream of “exiting” traditional nations by establishing new polities in cyberspace that eventually occupy physical space.
But despite the sheen of novelty afforded by the network state’s appeal to cutting-edge technologies, it is ultimately a recapitulation of the same old mistake. Recognizing that mistake, and resisting its understandable appeal, will be necessary if technologists are to develop a healthier relationship to politics and charter a better path forward.
A Country in the Cloud
The network state has as its basis the modern world’s digital infrastructure. Social networks can unite people on opposite sides of the globe, who have more in common with citizens of other nations than with their fellow countrymen. That dispersed sense of community and purpose can form a collective identity stronger than anything offered by the real world of political borders.
Balaji (who is universally referred to by his first name) argues that these digital communities, united by shared passions and interests, could form the basis of new states. A “startup society” can gradually attract believers to its collective vision, amassing capital and taking on the functions of everyday life through technologies unattached to a particular nation — conducting business through Bitcoin, for example, rather than through fiat currency.5 This network could eventually acquire a nation’s more abstract features, such as cultural norms and even a full-fledged social contract written into the laws of code.
Most of all, such a digital network can attract adherents to the “One Commandment” — the foundational principle or goal that all participants share.6 Because each network state is not tied down by the quirks of physical states, such as geographic location and language, its One Commandment can be as specific as its adherents want. And because people can opt in, everyone will be fully committed.
A social network could come to resemble genuine states to varying degrees. The least state-like example is a “digital network union,” whereby adherents form a quasi-guild that gives members some tangible benefit but which otherwise has no physical presence.7 Next is the “physical network archipelago,” which requires some physical footprint but no formal diplomatic recognition, because the community’s One Commandment is not at odds with any existing nation’s laws or norms.8 For example, a community committed to “Keto-Kosher” — both the ketogenic diet and Judaism’s dietary laws — could organize online and collectively purchase property, such as a group home, a cul-de-sac, or even a small town, in which to practice its lifestyle together.9 Because Keto-Kosher isn’t illegal anywhere, and at worst is seen as just an odd lifestyle choice, there is little reason for an existing state to oppose such collective micro-governance.
The most radical form of Balaji’s proposal is a “recognized network state,” which has attained diplomatic recognition from conventional states.10 Balaji’s example is a transhumanist, post-FDA, medical sovereignty zone in which members can buy or sell medical products freely.11 A sanctuary network state for biomedicine, free from the Food and Drug Administration’s regulations, would need positive legal recognition from an existing state, or at least a commitment that the existing state will not enforce existing laws against members of the network state. From there, a transhumanist network state could set up physical colonies around the world networked together by digital infrastructure, allowing them to live out their One Commandment in peace.
Balaji is clear that the model for these network states should be the tech industry, not just in that new technologies make a network state technically possible, but in that tech’s ethos should drive the creation of new states. His argument is full of analogies between states and startups; he calls for “a society run by Silicon Valley”; and he even calls the network state “Society-as-a-Service, the new SaaS.”12 Statesmen and military thinkers during the Cold War spoke of “push-button warfare”; Balaji goes one step further and proposes what we could call “push-button diplomacy.”
The beauty of the network state is its voluntary nature. Whereas conventional states rely on coercion to force people into citizenship, network states allow people to opt into whichever political community best fits their principles, leaving behind those who prefer the nations of old. As Balaji puts it in his lecture, “If you like your country, you can keep it.”13
The Limits of the Network
You can keep your country if you like it, but can we really just renounce political geography and flip our allegiance? Balaji seems to envision a world where changing one’s political affiliation is as simple as creating a new Minecraft server, with a new digital Eden always a few clicks away. But true “exit” as Balaji envisions — the ability to leave an existing system entirely and build or join a new one with its own governance, culture, and infrastructure — is a mirage.
That last piece, infrastructure, is notably missing from Balaji’s argument. In the 260 pages of his book The Network State, Balaji mentions “infrastructure” only three times, and then only to discuss Chinese history and diplomacy. But infrastructure forms the nuts and bolts of human life. With all the trappings of the modern world, it can be easy to forget how fragile, centralized, and controllable our infrastructure really is. To be truly independent, a network state would need to build its own power plants, run its own telecommunications cables, host its own servers, manufacture its own computers, and write its own software. The internet isn’t independent from the industrial world; it is inextricably reliant upon it.
More importantly, even a country in the cloud must establish a sustainable supply chain for food, water, and shelter, and it must do so within the rules and regulations for the nation in which it exists. These supply chains are heavily regulated and heavily interdependent. Consider electricity production and transmission. There are only 66 balancing authorities that operate the U.S. electrical grid, and only three regional interconnection grids, all of which are heavily regulated. Each regional grid is complexly intra- and interconnected; a single tree branch tripping a transmission line can disrupt service for 50 million people.14
Network state advocates might retort that this is exactly the kind of centralized mess they are trying to avoid — “Just build your own micro-grid!” But government regulators would have the last laugh. As soon as any network state’s electricity grid spanned multiple properties, delivered service to a third party, wanted a backup in case of unexpected outages, or chose to sell excess power back to the grid, it would be required to interconnect with the main grid. Even if a network state somehow managed to build an islanded, behind-the-meter network that did not interconnect, states have strict rules over where and how individuals can build micro-grids. And all this doesn’t even begin to consider the environmental laws with which a network state would need to comply.
The State of Atoms vs. the State of Bits
Infrastructure poses enough of a problem, but there is an even deeper challenge to digital utopia. History shows that states do not tolerate breakaway societies challenging their authority. Once an “exit” project gains even modest traction, nearby governments react with legal force or even military action to quash it.
When libertarians tried to live outside any nation’s jurisdiction by building a floating seastead off the coast of Thailand, the Thai Navy raided and seized it, charging the occupants with violating Thailand’s sovereignty.15 In another case, a group attempted to found the “Republic of Minerva” in the Pacific, only to be forcibly removed by Tonga’s troops within weeks.16 Whenever and wherever people have attempted to quit society and build anew, existing political authorities have stood in their way. Indeed, even the godfather of seasteading, Peter Thiel, observed last year that escaping the state is only getting more difficult.17
Even in countries that generally respect the right of free association, insular communities exist only as long as the government allows them to, and as long as they stay within the bounds of their political superiors. Go ask the Branch Davidians, the Brotherhood of Eternal Love, or the Fundamentalist Church of Jesus Christ of Latter-Day Saints, each of which was broken up by American law enforcement after running afoul of the law.
Official recognition of a people’s independence by the state can be just as tenuous. Consider the story of the Apache. In the 1852 Treaty of Santa Fe, the U.S. government acknowledged Apache sovereignty and promised protection, supplies, trading posts, and a halt of expansion into Apache territory. When the government didn’t live up to its end of the deal, open warfare began a decade later and lasted until 1872, after which the government signed another treaty establishing a reservation in Arizona. But it too was dissolved soon enough, and in 1876 the Apache were forcibly relocated. It wasn’t until 2014 that some Apache received rights and official recognition promised more than a century ago. Even today, to put it cynically, the only thing preventing the U.S. government from once again rolling cavalry across the Southwest and forcibly relocating the Apache is a promise and a piece of paper.
No individual or community, no matter how large or online, can fully escape politics. Any network state that attracted a significant population or wealth would face the usual fate: crackdown or co-option by threatened nations. Whatever legitimacy the state of bits has on the world stage is ultimately dependent on the clemency of the state of atoms.
Balaji’s Fallacy
Balaji’s error is not a new one. The dream of tech exit has been a false hope for the whole history of the digital age, luring technologists away from governing in the real world and into a fantasy of political escape.
Early computing visionaries saw in the computer the ultimate escape hatch from society. Countercultural hero Stewart Brand boasted that the “freaks who design computer science” would defeat the “rich and powerful institutions.”18 Engineer Tim May’s Crypto Anarchist Manifesto promised that “[t]he State will of course try to slow or halt the spread of [the computer].… But this will not halt the spread of crypto anarchy.”19 Activist John Perry Barlow’s Declaration of the Independence of Cyberspace declared, “Governments of the Industrial World … I declare the global social space we are building to be naturally independent of the tyrannies you seek to impose on us.”20 For the cyber-hippies, the computer would let us all turn on, tune in, and drop out of the state. Books such as Walter B. Wriston’s The Twilight of Sovereignty and William Rees-Mogg and James Dale Davidson’s The Sovereign Individual similarly prophesied that digital communications would thwart yesteryear’s physical tyrants and free the individual revolutionaries, computer in hand.
But it’s never worked out that way. Instead, the internet has led to tremendous centralization, homogenization and, in some cases, oppression. In China, Russia, and Iran, it has enabled surveillance and subjugation beyond the dreams of any pre-internet despot. To today’s victims of digital authoritarianism, the promise that the internet would evade the state and cause a thousand digital flowers to bloom must seem a bad joke.
Voice over Exit
The “opt-out” fantasy has repeatedly collided with reality. Groups attempting a strategic withdrawal from society have found themselves frustrated by the physical realities of modern infrastructure, and those attempting to escape state control have been either crushed or absorbed by prevailing powers. James Madison famously wrote in Federalist 48 that “parchment barriers” and the promises of existing states to respect citizens’ autonomy were “not a sufficient guard against those encroachments which lead to [tyranny].”21 The network state, in promising a digital end-run around these historical, technological, and political realities, overlooks the fundamental power imbalances and practical dependencies that make its dream unworkable.
What, then, are the malcontents to do? If it is impossible to build a new country in the cloud, those who are dissatisfied with their status quo may slip into nihilistic apathy. But another approach is possible. Instead of dreaming of becoming high-tech Robinson Crusoes, technologists should strive to serve the political communities they already belong to.
We are seeing today a renewed recognition of this lesson in other corners of the tech industry. Whether it’s former PayPal COO David Sacks’s work in the White House, Y Combinator CEO Garry Tan’s crusade to reform San Francisco’s governance, or J.D. Vance’s departure from venture capital to run for the U.S. Senate and, later, the vice presidency, technologists are starting to appreciate the merits of “voice,” to use Albert O. Hirschman’s terminology, over “exit.”22 Balaji himself courted President Trump in an effort to reform medical and pharmaceutical policy and, reportedly, to throw his hat in the ring to lead the FDA.23
At some level, it seems, nearly everyone understands that, when it comes to politics, if you can’t beat ‘em, join ‘em. If you can’t exit society, you might as well join the generational endeavor to improve it. After all, what sort of polity would a network state be if anytime the going got tough, its members blasted off to some other message board, planted a new digital flag, and declared it Year Zero in the world of ones and zeros yet again?
- Jack Goldsmith and Tim Wu, Who Controls the Internet? Illusions of a Borderless World (Oxford University Press, 2006), 6. ↩︎
- Ibid., 10. ↩︎
- Balaji Srinivasan, The Network State (published by the author, 2022), 9. ↩︎
- “Balaji Srinivasan at Startup School 2013,” YouTube, 2013; “The Network State Conference,” Network School, 2025. ↩︎
- Balaji Srinivasan, The Network State (published by the author, 2022), 244. ↩︎
- Ibid., 136. ↩︎
- Ibid., 137. ↩︎
- Ibid. ↩︎
- Ibid., 138–139. ↩︎
- Ibid., 140. ↩︎
- Ibid. ↩︎
- Ibid., 28; “Balaji Srinivasan at Startup School 2013,” YouTube, 2013. ↩︎
- Ibid. ↩︎
- “August 2003 Blackout,” U.S. Department of Energy. ↩︎
- Brian Doherty, “Thai Government Takes Over Seastead Near the Thai Coast,” Reason, April 15, 2019. ↩︎
- Raymond Craib, “The Brief Life and Watery Death of a ’70s Libertarian Micronation,” Slate, May 21, 2022. ↩︎
- “Peter Thiel on Political Theology,” Conversations with Tyler, April 17, 2024. ↩︎
- Quoted in Walter Isaacson, The Innovators: How a Group of Hackers, Geniuses, and Geeks Created the Digital Revolution (Simon & Schuster, 2014), 269. ↩︎
- Tim May, The Crypto Anarchist Manifesto, Satoshi Nakamoto Institute, 1988. ↩︎
- John Perry Barlow, A Declaration of the Independence of Cyberspace, Electronic Frontier Foundation, 1996. ↩︎
- James Madison, “Federalist: No. 48,” 1788, accessed via The Avalon Project, May 17, 2025. ↩︎
- Albert O. Hirschman, Exit, Voice, and Loyalty: Responses to Decline in Firms, Organizations, and States (Harvard University Press, 1970). ↩︎
- Thomas M. Burton, “Donald Trump Looking Beyond Traditional Medical Experts for FDA Commissioner,” The Wall Street Journal, January 13, 2017. ↩︎