How Jack Dorsey Built an App Governments Couldn't Monitor. And Why This Was the Problem.
By Creatives Takeover Editorial Team · October 6, 2026
The feature users loved was the exact feature governments couldn't allow.
In August 2011, a British Member of Parliament stood up and asked BlackBerry's parent company to simply turn off its messaging service during the London riots. The company considered it, then agreed to cooperate with police instead, and was promptly hacked by the very people it had been trying to track. Fifteen years later, Jack Dorsey built an app specifically so that conversation could never happen again, because there would be no company left to ask.
A Government's Oldest Trick Stopped Working
For as long as messaging apps have existed, governments trying to manage a crisis have relied on one basic fact: somewhere, a company runs the servers. That company has an office, a legal team, and a business it doesn't want shut down, which means it can eventually be pressured, sued, or legally compelled to hand over whatever a government wants to see. This is exactly what happened to BlackBerry in 2011, when its parent company RIM faced direct political pressure to either suspend its messenger service or cooperate with London police chasing rioters who'd used it to organize. It's exactly what happened to Telegram in Russia in 2018, when a Moscow court ordered the app blocked outright after its founder refused, repeatedly, to hand security services the encryption keys needed to read users' messages. Iran ran the same playbook against Telegram that same year, citing national security during a wave of anti-government protests, even though the app's own founder said he'd refused specifically to shut down channels being used for peaceful organizing.
In every one of those cases, the fight was really between a government and a company. The company had leverage, headquarters, legal exposure, a reputation to protect, and in Telegram's case, a founder willing to publicly resist for years. That leverage is also exactly what gave governments a lever to pull. Jack Dorsey, when he built Bitchat in July 2025, built something with no equivalent lever anywhere in it.
The App With Nobody Behind the Counter
Dorsey put Bitchat together over a single weekend, using an AI coding tool, and published a short technical paper laying out exactly what it would and wouldn't do. There's no sign up, because there's no account system to sign up for. There's no phone number attached to a user, no password to reset, and no company server quietly holding a copy of every conversation in case someone eventually asks for it. Messages move directly from one phone to the next over Bluetooth, hopping across nearby devices until they reach whoever they're addressed to, the same basic idea as a two-way radio, just automatic and encrypted along the way.
This wasn't an accident he discovered after the fact. Dorsey has spent years talking about what he calls "protocols, not platforms," the idea that any single company sitting in the middle of a conversation is a weak point rather than a convenience, because that company can always eventually be pressured into cooperating, the way BlackBerry and Telegram both were, one way or another. With Bitchat, he removed the weak point entirely by making sure there was no company there to pressure in the first place.
Why the Old Playbook Breaks Completely
Put the two eras side by side and the shift becomes obvious. With BlackBerry in 2011, the UK government's actual request was aimed at a company: turn off the service, or help us find who's using it. With Telegram in Russia and Iran, the demand was just as direct: hand over the encryption keys, or we'll block the app nationwide. Both demands assumed, correctly, that there was an organization on the other end capable of complying, refusing, or negotiating a middle ground.
Bitchat has no equivalent recipient for that kind of demand. There's no company to order into cooperation, because nobody runs the network, it's simply made up of whoever happens to have the app installed nearby at any given moment. There's no central server to seize or subpoena, because messages pass hop to hop between phones and are designed to vanish after delivery unless a user specifically chooses to keep them. And there's no identity attached to trace a message back to a person, even in the rare case someone could intercept one mid-hop. Governments reacting to Bitchat haven't been negotiating with a company that might eventually fold the way Telegram eventually did in some markets. They've been discovering, one country at a time, that there's simply nobody on the other end of the line to negotiate with at all.
Four Governments, One Conclusion, Almost No Overlap Otherwise
That discovery has played out in remarkably similar ways across countries that share almost nothing else in common. In Nepal, in September 2025, young people protesting government corruption turned to Bitchat by the tens of thousands within days, right after officials imposed a temporary social media ban. In Uganda, around a contested election, an opposition figure publicly endorsed the app as a way to stay connected through government blackouts, and the national telecom regulator publicly threatened to shut it down, a threat a developer connected to the project brushed off almost immediately, pointing out there was no server to switch off.
China went further than a threat. In February 2026, regulators there forced Apple to pull Bitchat from the country's app store entirely, citing rules against unauthorized social mobilization. And in India, during protests in Delhi tied to an exam scandal in mid-2026, officials suspended mobile internet in parts of the city, pushing protesters toward Bitchat, after which the country's cybercrime agency ordered the app's source code removed from GitHub, stating outright that its design "significantly impedes lawful interception, attribution, and investigation by law enforcement agencies." By October, that same government had gone a step further and had the app pulled from its app store entirely. Dorsey, responding publicly each time, didn't dispute any of it. He simply said the relevant government didn't like a technology like this existing, and wanted it gone.
Four governments. A fragile democracy managing youth unrest, an East African state mid-election, the world's largest authoritarian government, and the world's largest democracy. Almost nothing else links them politically. All four reached the same operational conclusion about the same piece of software within about a year of its release.
What Telegram Could Still Do That Bitchat Can't Be Made to Do
There's a detail in Telegram's history that makes the comparison sharper rather than softer. Even after years of standoffs, Telegram's founder Pavel Durov still had a choice to make each time a government came calling, comply, negotiate, or accept the app getting blocked in that market while continuing to operate everywhere else. That's a real decision a real person made repeatedly, under real pressure, and it's part of why Telegram has such a long, dramatic history of specific confrontations with specific governments stretching back years.
Bitchat was built so that no equivalent decision point exists anywhere in the system. Nobody can be called into a negotiating room, because the app doesn't route through anyone capable of making a decision on its users' behalf once it's running on their phones. That's a meaningfully different kind of resistance than a founder refusing to hand over a key. It's closer to removing the lock entirely, so there's no key to ask for in the first place.
What This Actually Means for a Founder Building Right Now
The temptation is to draw a simple lesson here, something like "build in resilience against censorship," but that undersells what actually separates this case from the decades of encrypted messaging apps that came before it. The real shift is that Dorsey didn't just make Bitchat harder to monitor than existing options. He removed the one thing every previous government strategy, from BlackBerry to Telegram, had depended on to apply pressure at all: an organization capable of receiving that pressure and eventually responding to it.
If you're building something where the entire value to users comes from the absence of a central authority, understand that you're not just building a more private product than what's already out there. You're removing the exact mechanism every government has historically used to manage a technology it doesn't like, which means the usual playbook of compliance requests, local offices, and negotiated settlements simply won't apply to you, for better and for worse. That's a fundamentally different risk profile than building a product a government merely dislikes, and it's worth knowing which one you're actually building before you find out from a takedown order.
Five Things Worth Taking From This
Removing the company doesn't just remove a risk, it removes the entire category of response governments have relied on for a decade. BlackBerry and Telegram both eventually had to make a decision when pressured. Bitchat was built so nobody is in a position to make that decision at all.
A government's reaction to a new technology often depends less on its politics and more on what the technology structurally allows. Nepal, Uganda, China and India have almost nothing in common politically, yet all four reached the same conclusion about the same app within roughly a year.
History offers a useful before-and-after here, not just a list of comparisons. Telegram's years of standoffs with Russia and Iran show what the old fight looked like, a company deciding whether to comply. Bitchat shows what happens once that company is designed out of existence.
A feature that resists monitoring becomes most valuable at the exact moment monitoring matters most to the people trying to do it. Every surge in Bitchat's download numbers lines up with a shutdown, a blackout, or an election, which is also exactly when governments have the least patience for a technology they can't reach.
Some products can't be regulated into compliance, only removed outright. A government asking Telegram for an encryption key was asking for something that could technically be handed over. A government asking for Bitchat's equivalent has nothing to ask for, which is why removal, not negotiation, has been the response every time.
Dorsey had already watched what happened to the apps before his. He'd seen BlackBerry weigh cooperating with police against its users' trust, and watched Telegram spend years in court over a set of encryption keys it could, in theory, have handed over. Bitchat reads less like an attempt to win that fight than a decision to make sure it never had to be fought again, by building something with no one left standing in the room when the question finally gets asked.