Radar Chat: Bitcoin Payments Inside Signal

Vik Sharma and Seth For Privacy join to lay out what Radar actually is, how the Spark protocol makes self-custodial Lightning feel effortless, and why the digital panopticon makes this the most important freedom tech launch in years.

Marty Bent

July 19, 202615 min read

Vik Sharma and Seth For Privacy on the TFTC podcast discussing Radar Chat, a Signal fork with Bitcoin Lightning payments

↓ Jump to the video and timestamps

Matt Odell and I have been talking about the need for Bitcoin payments inside Signal for roughly eight years. Rabbit Hole Recap episode after episode, the same refrain: why can't freedom money flow through the same app we use for freedom speech? This week, Vik Sharma from Cake Wallet and Seth For Privacy sat down with me to walk through Radar Chat, a Signal fork that finally answers that question with a real product that real people can download today.

The launch date was July 7th. That timing against the current backdrop is not coincidental. The EU's Chat Control proposal is still advancing. VPN restrictions are being discussed in the UK. Facial recognition requirements are being floated for vehicles across Europe. The digital panopticon is being erected in real time, and it falls on people in the freedom tech space to build the tools that route around it. Radar is exactly that kind of tool, and it is the most important thing to ship in this space in a long time.

What struck me most about the conversation is how different this moment feels from even 18 months ago. The UX argument against freedom tech has always been the killer: great in theory, unusable in practice. That objection is running out of runway fast.

Key takeaways

Radar is a Signal fork with self-custodial Bitcoin Lightning payments built in. It runs on the Spark protocol via the Breez SDK, your seed phrase is encrypted into your Signal profile backup, and it is fully interoperable with your existing Signal chats from day one.
The addressable pool is tens of millions of Signal monthly active users (estimates commonly cite around 70 million), most of whom are not Bitcoiners. That is the actual opportunity, not converting people who already have wallets.
Convenience parity with freedom tech is no longer hypothetical. Signal cracked the UX problem for private messaging. Radar is betting that the same trick works for private payments.
Privacy has not been chosen out of people, it has been conditioned out of them. The early internet was not surveilled by default. Corporations built that norm, and the tools to break it are now genuinely good enough to compete on experience.
Tucker Carlson claimed Signal was intercepted, but the likely culprit is device compromise. Seth's read, and mine, is that Signal's protocol held and the endpoint is where things went wrong.
The agentic AI future could make full Bitcoin self-custody accessible to people who have no idea what inbound liquidity means. An agent that manages your Lightning node and executes voice-command payments removes the last meaningful UX objection to sovereign Bitcoin ownership.

Why Radar Launched When It Did

The timing matters. July 7th, and the news cycle around it reads like a brief for why this product has to exist.

The EU's chat control proposal, which would compel providers to scan private messages for illegal content, has been grinding through the legislative process for years. The UK has been discussing restrictions on VPN services. Reports have circulated about facial recognition requirements being considered for vehicles across Europe. Every one of these is an attempt to close off the spaces where people can speak and transact without being watched.

As I put it in the conversation: "the digital panopticon is being erected around us and it's incumbent upon us in the Freedom Tech space to build the tools to route around that."

Vik's origin story for Radar runs through two different observations. The first was community demand. Odell has been beating this drum for years, and he is not alone. The second came from his trips to China, watching everyone use WeChat for payments at the point of sale without ever switching apps. If you are at a cafe and need to pay, you do not open a separate app. You are already in your messaging app, and the money just moves. That is the experience Radar is trying to replicate, with the surveillance architecture of WeChat stripped out entirely.

The technical readiness piece is what made now the right moment. Vik tried to put Lightning into Cake Wallet years ago and killed the feature because the user experience was genuinely bad. Then Breez built an SDK on top of the Spark protocol, Cake shipped it earlier this year, and it worked. That success gave them the confidence to port it into Signal.

Vik's framing of the bigger trend is worth sitting with: "Rather than making things for just bitcoiners, make things for regular folks and put bitcoin in there." Jack Dorsey did it with the Square terminal, and there has even been a Bitcoin for Signal campaign pushing the same idea. Radar is doing it with messaging.

What Radar Actually Is

At its core, Radar is a fork of Signal. It is fully interoperable with your existing Signal chats. You back up your Signal profile, restore it inside Radar, and your entire chat history comes with you. Vik and Seth have been running it as their primary Signal client for four months. The person on the other end cannot tell you switched.

The payment layer runs on the Spark protocol via the Breez SDK. Self-custodial Lightning, meaning you hold the keys. Your seed phrase is encrypted and backed up directly into your Signal profile, so if you lose your phone and log into a new device, your sats come back automatically. You do not manage a separate seed phrase. You do not worry about inbound liquidity. You do not need to know what any of that means.

Seth's explanation of why Signal and not something else is the right answer to the obvious question. WhatsApp and Telegram use privacy as a marketing veneer, his words. "They kind of use that as a marketing veneer, something to put on what they're actually building, but it's not core to what they're doing." Signal is different. The privacy is structural, not promotional. And the messaging protocol is battle-tested enough that you do not need to reinvent it. Build on the gold standard, then add the freedom money layer.

The total addressable audience, Signal's monthly active user base in the tens of millions (estimates commonly cite around 70 million), matters because the vast majority of those people are not Bitcoiners. They are just people who want a decent cross-platform messenger. Radar is not trying to build another wallet for people who already have wallets. It is trying to put Bitcoin in front of people who will not notice it is Bitcoin until they have already used it.

The Convenience Parity Problem, Finally Solved

The historical knock on freedom tech is not ideology. Nobody argues that privacy is bad. The argument has always been convenience. The walled garden wins because it is easy, and the alternative is too painful to bother with.

Seth made the point about Signal's own user base better than I could: he has gotten his entire family onto Signal over the years. Not one of them understands what the Triple Ratchet protocol is (the post-quantum upgrade to Signal's older Double Ratchet) or why it matters. They are on Signal because Seth is on Signal and because it works. That is the model. You abstract the complexity and compete on the experience.

Vibe coding is changing the power dynamic here in a way that was not true even three years ago. Seth built the Radar website himself. He is not a designer or a developer. But he had a strong sense of what he wanted, good brand foundations from their head of brand, and AI tools that could bridge the gap. That website exists because the barrier to producing a polished, approachable product has collapsed.

Vik made the open-source point that often gets overlooked in these conversations. Cake Wallet's code is public. If the company disappeared tomorrow, you could download the code, feed it into an AI tool, and build yourself a functional wallet. Three years ago that was a theoretical comfort. Today it is a practical one. The government cannot dock code that is already out in the world, and the tools to make something usable out of that code are now within reach of people who would never have called themselves developers.

Privacy Is Conditioning, Not Preference

Here is the riff I keep coming back to after this conversation.

When you mail a letter, you lick the envelope and seal it. That is your privacy technology. It is a thin layer of dried glue. And yet we as a society treat the violation of that seal as serious enough to make it a federal offense. Open someone else's mail and you have committed a crime.

Now think about how we have come to treat digital communication. The assumption for most people is that privacy online is either not possible or not worth pursuing. That assumption is not a preference. It is conditioning. Seth put it cleanly: "The early Internet days were not like this, but quickly it got taken over and the realization that the data that's available in how people are using the Internet is immensely valuable. And so we got conditioned by all of these corporations that were building and abusing our data."

The building blocks were always there. Cryptography and the cypherpunk tradition are foundational to how the internet actually works. What corporations did was build surveillance on top of those foundations and convince us that was just how things were. It was not. It is not.

Seth's read on where sentiment has moved is encouraging. The "I have nothing to hide" retort is dying. The question people are actually asking now is whether the tools exist to do something about it. And the answer, for the first time in a long time, is yes without the asterisk.

Apple is running privacy as a central marketing message not because Apple is suddenly ideologically committed to cypherpunk values but because people are actually buying based on it. That is a market signal. The demand is real.

Tucker Carlson, Signal, and What OpSec Actually Means

Tucker made a lot of noise claiming Signal was compromised after his conversations with Russian officials apparently surfaced. I'll say plainly what I said on tape: Tucker did a very poor job of besmirching Signal when the problem, by the read of most security researchers, almost certainly had nothing to do with Signal.

Seth's breakdown is the right one. "Pwning your device is going to be drastically easier than pwning signal at this point." The Signal protocol itself has remained cryptographically sound under enormous scrutiny. The far likelier explanation is a device-level compromise of the sort Pegasus-style spyware enables. If you are a high-profile target, a journalist, an activist, a senior government official, you should assume your device is compromised. Signal cannot save you from a device-level intrusion. Lockdown mode on iOS exists for exactly this reason, and if you are in that threat category, you should be using every hardening tool available.

There is also the version of this problem that has nothing to do with sophisticated attacks. The government officials in the Signalgate episode who accidentally added a journalist to a Signal group chat did not have their devices compromised. They just added the wrong person. Signal protects the channel. It cannot protect you from yourself.

The operational security lesson is that privacy tools are a stack. Signal is one layer. Device security is another. Who you add to a chat is a third. Confusing a failure in one layer for a failure in another is not just wrong, it actively misleads people about where the real risks are.

Spark vs. Ark vs. Lightning: Why the Semantic Wars Miss the Point

I spent two days at a Bitcoin conference a while back interviewing people on both sides of the debate over Bitcoin Core v30 lifting the OP_RETURN datacarrier limit. By the end of it, I had concluded pretty firmly that continuing to engage was not worth the energy. The same principle applies to the Spark versus Ark arguments that have been circulating.

I do not wade into those arguments anymore. Let's just see on the field.

Seth gave the honest technical picture: Radar, built on Spark, does not currently support seamless unilateral exit. That is a real limitation worth knowing about, and the Breez team is actively working to fix it. Calling that out is fair. What is not productive is the trench warfare version of these debates where every party is trying to grind down the other side rather than building.

The thing people miss when they treat Spark and Ark as zero-sum is that they are both open protocols operating over Lightning. The switching cost for a user is essentially nothing. If you decide you want to move from a Spark-backed wallet to an Ark-backed one, you send your sats over Lightning and you are done. No new account, no friction, no walled garden holding you in.

Seth put it well: "One of the beautiful parts of this is they have to really aggressively compete for the user because the switching cost is essentially nothing." That competition makes both better. The mud-throwing does not. The real beneficiary of the protocol wars is neither Spark nor Ark. It is the closed, custodial, surveilled alternatives that neither of them are.

Lightspark and David Marcus have taken a lot of heat from the more ideologically committed corner of the Bitcoin space, but I think the honest read is that they have pulled the field toward building things that normal people actually want to use. That is net positive. We can hold them accountable on the things that matter without treating every pragmatic trade-off as a betrayal.

Stablecoins, Agentic AI, and the Road to 10 Million Users

Vik threw a number out when I asked what success looks like: 10 percent of Signal's user base, which by his math puts you at around 10 million users. Cake Wallet sits at roughly 2 million. The distinction is that Cake is a cryptocurrency wallet. You download it because you want a wallet. Radar is not that. The opportunity is different in kind, not just in scale.

Stablecoins are coming to Radar. Seth, in a line I want to put on a wall, said: "I hate them. I hate everything that a stablecoin stands for and that it reinforces the legacy fiat system. They're almost always easily surveilled. Like they're in every way they're the opposite of freedom money." And then he explained why they are going on the roadmap anyway.

Because when you talk to people on the ground in countries that actually need freedom money, they are using stablecoins. They need the stability. They are not ready for Bitcoin volatility. So the play is to put stablecoins on Bitcoin rails, specifically on Spark, let people choose their balance denomination, and make the swap to hard money a single tap away. The goal is to have that tap be the obvious move the moment a user's stablecoin balance gets frozen, censored, or otherwise fails them. You plant the rail, and you wait for the pain to do the rest of the educating.

The group chat features on the roadmap feel like the killer use case that will drive real adoption. Splitting a bill across a group, running fundraising directly inside the chat without routing through a third-party platform, those are things people actually want to do with the people they are already talking to.

The agentic angle I raised in the conversation is the one I keep thinking about. The form factor of how people interact with computers is changing faster than most product teams are accounting for. I can already see a world where you just tell your AI agent that Seth covered dinner and ask it to send $20 in sats, and the agent handles the Lightning node, the liquidity, the payment, everything, without you touching a single setting. Seth confirmed the agent could theoretically run a full Lightning node and manage inbound liquidity entirely autonomously. That removes the last serious UX objection to full self-custodial sovereignty, which is that it requires technical knowledge most people do not have and are not going to acquire.

That world is closer than most people think. And when it arrives, the only question is whether the rails underneath are Bitcoin or something else.

About Vik Sharma and Seth For Privacy

Vik Sharma is the founder and CEO of both Cake Wallet and Radar, two separate companies that share his leadership. Cake Wallet is a self-custodial, open-source wallet with millions of users globally. He has been a builder in the freedom tech and privacy space for years and is one of the driving forces behind Radar Chat's development.

Seth For Privacy is a pseudonymous privacy educator, writer, and technologist, the COO of Cake Wallet, and the host of the Opt Out podcast. He led Cake's Lightning work, co-built Radar, and has been a vocal proponent of building privacy tools with user experience that can compete with surveilled alternatives. He has been running Radar as his primary Signal client since before launch.

Sources mentioned

Radar Chat: the product itself, where to download and learn more
Cake Wallet: Vik Sharma's open-source wallet, where the Breez Spark SDK was first deployed
Signal: the open-source messaging app Radar forks and stays interoperable with
Spark: the Bitcoin Layer 2 powering Radar's self-custodial payments
Breez SDK: the SDK underlying Radar's Spark-based payment layer
Signal's Triple Ratchet upgrade: Signal's post-quantum successor to the Double Ratchet
Ark protocol: the other open protocol in the Spark versus Ark debate
Lightspark: David Marcus's company behind Spark
Matt Odell: who has pushed for Bitcoin inside Signal for years
EU Chat Control: the message-scanning mandate that frames the launch
Signalgate: the group chat where officials accidentally added a journalist
Pegasus spyware: the device-level attack vector behind the contested Tucker Carlson claim

Watch the conversation

Timestamps

0:07 - Intro / Bitcoin as safe haven
1:08 - Why Radar, why now: surveillance backdrop and the July 7 launch
2:17 - Vik's origin story: WeChat, community demand, and the tech maturity moment
4:58 - How Radar works: Signal fork, Spark protocol, self-custodial Lightning
9:19 - The bigger trend: embed Bitcoin into existing tools, not new ones for Bitcoiners
10:02 - Signal vs. WhatsApp vs. Telegram, and why not stablecoins from day one
19:30 - Vibe coding, open source resilience, and the UX parity moment
33:00 - The envelope riff: privacy conditioning and the digital double standard
43:00 - Tucker Carlson, Signal, and what device security actually means
52:00 - Spark vs. Ark semantic wars and why I stopped engaging
1:02:00 - Stablecoins, group chat features, and the agentic AI payments future

Frequently Asked Questions

Radar is a fork of the open-source Signal application with self-custodial Bitcoin Lightning payments built in via the Spark protocol. It is fully interoperable with Signal, meaning you can migrate your existing chat history and continue messaging Signal users without them knowing you switched apps. The difference is that Radar also lets you send and receive Bitcoin directly inside the same interface you use to message people.

Yes. Your seed phrase is generated on your device and backed up encrypted into your Signal profile. If you lose your phone and log into a new device, your funds come back automatically. Seth describes it as still self-custodial under the Spark trust model, though it is worth understanding that Spark does have trade-offs relative to running a full Lightning node, including current limitations on seamless unilateral exit that the Breez team is working to resolve.

Yes. The process is to do a full backup in Signal first, save the recovery key it gives you, and then restore that backup inside Radar. Your chats come with you. The team is adding a prompt to the next update to make sure people complete the Signal backup step before proceeding, since Radar will replace Signal as the primary app on that device.

Spark is a Layer 2 protocol built on top of Bitcoin that enables self-custodial Lightning-style payments without requiring users to manage channels or inbound liquidity. The Breez SDK provides the implementation that Radar runs on. The main current trade-off is that unilateral exit, meaning your ability to pull funds on-chain without counterparty cooperation, is not yet seamless inside Radar directly. You can do it via separate tooling by importing your seed phrase, and the Breez team is actively building a smoother path.

Signal is the gold standard for private messaging protocols. Building a new messaging protocol from scratch introduces enormous surface area for things to go wrong, and it means starting a network from zero users. Signal already has an estimated tens of millions of monthly active users (commonly cited around 70 million) and the gold standard privacy stack. Forking it let the Radar team focus on the payments layer without reinventing the messaging wheel, and it means day one users can message their existing Signal contacts without asking anyone to switch platforms.

No. The Signal protocol itself has remained cryptographically sound. What Tucker almost certainly experienced was a device-level compromise, possibly via sophisticated spyware, not a failure of Signal's encryption. Seth's point is the right one: if you are a high-profile target, assume your device is compromised and act accordingly. Signal cannot protect you from a pwned endpoint. That is an OpSec stack problem, not a Signal problem.

Stablecoin support is on the roadmap, but only stablecoins that live on Bitcoin rails, specifically on Spark. Seth is explicit that stablecoins are ideologically the opposite of freedom money. The reasoning for including them is pragmatic: a large portion of the people who most need censorship-resistant payments are currently using stablecoins because they cannot absorb Bitcoin's volatility. The goal is to give them Bitcoin rails underneath, make the swap to hard money a single tap, and let the failures of censored stablecoins do the work of convincing people to move.