Copper Founder: Building a Billion Dollar Company & Fixing Crypto Custody
- Dmitry Tokarev says Copper raised $300M in the ZIRP era, reaching a latest valuation of $1.25B, because quietly building a custody business was no longer enough: awareness creates trust. He had to spend to be “everywhere that's humanly possible,” including an Iceland commercial with Rebecca Ferguson so expensive he avoided Dune: Part Two to not remember the cost. The harder part was cutting burn after growing from 36 to 236 people between October 2021 and October 2022.
- His insider read on FTX is that it was not exclusively engineered for mass fraud — “he just lost control” as the company scaled. Copper signed documents to integrate FTX into ClearLoop in July 2022, then chased Sam Bankman-Fried while his team said he had 130 direct reports. A client-funds/FTT leverage problem that may initially have been $1M–$2M grew to $1B, $2B, and $5B because nobody could stop it.
- Bron/Brown is his “gen-three” self-custody wallet: institutional MPC, with no single private key, democratized for individuals. It is at about 1,000 users and $1B in assets, mostly family offices, executives, and founders. Gen one kept keys in memory, gen two put them on devices, and gen three addresses recovery, inheritance, hidden wallets, and programmable policies such as requiring approval above $100K in 24-hour volume or imposing a 48-hour delay above $1M.
- The threat landscape is part of the pitch: Tokarev says physical attacks happen “weekly, that I’m aware of. Probably daily,” while social-engineering scams exploit leaked exchange data. Criminals target less-protected holders rather than highly visible, security-conscious figures. Three friends were scammed by bribed support staff; one transferred $100K in Bitcoin after callers supplied transaction details. He also cites a task force he thinks was called Operation Atlantic, involving the U.S. Secret Service, U.K. National Crime Agency, and Ontario Provincial Police, to fight DeFi phishing. Lost keys also destroy supply, as with what he thinks was Matthew Mellon’s roughly half-billion-dollar Ripple holding.
- The project has no equity entity: it uses a fixed-supply, 100% utility token with KPI-based unlocks tied to user growth rather than time. Tokarev expects the first time-based unlocks, if any, to be deep in the 2030s. He says the token model gives customer LTV immediately because the customer is already there, while skeptics can pay a subscription.
- AI has changed his scaling plan: the team stopped hiring after secure workflows with Claude Code became practical roughly three to six weeks earlier, and it remains under 40 people. His advice is to convince a customer to pay real dollars before fundraising. He is also “super bullish on salespeople again,” arguing corporate buyers need a person whose judgment they can rely on because they cannot “cover your ass with Claude.”
- Tokarev says he bought Bitcoin for the first time in two years when Iran was attacked and it rose slightly; he views roughly $56K realized price and fear and greed near 4/100 as bottom-related signals. He is bullish on Canton because Wall Street is writing DAML, owns some Zama for fully homomorphic encryption, and is considering decentralized protocols for perps, lending, trading, and options: Hyperliquid, Aster, Derive, Avax, Camino, and Jupiter. He calls Hyperliquid probably the most robust but does not expect a winner-takes-all market.
1. Institutions never saved crypto — crypto built the rails itself, and the question now repeats
- Tokarev's opening reframe from the Franklin Templeton booth at DAS: since 2019 the theme was “the institutions are coming,” but shops that went public, such as Circle and BitGo, or are preparing to do so, were already around seven years ago hoping institutions would rescue them. Instead, “they and all of us” built the industry. The same question now applies to traditional assets coming on-chain.
- His sizing argument: crypto's roughly $2.5T–$3T market cap is about Nvidia's market cap, while the entire current infrastructure services roughly that scale. He asks listeners to imagine traditional assets arriving on-chain and all of those assets having to live a “blockchain life.”
- He also sees a behavioral shift: people are less interested in having a manager manage their assets or buying an index. They increasingly want to understand market dynamics and choose single-asset or single-commodity exposure, including palladium, gold, or silver.
2. Copper's $300M total raise: forced marketing, four U.S. competitors, and 36→236 people
- Copper quietly built a business in its first years, but in the zero-interest-rate era it could not remain profitable yet unknown. In custody, “awareness creates trust,” so it raised $300M in total; the latest valuation was $1.25B. The money went toward marketing and being “everywhere that's humanly possible,” after which competitors raised more and Copper had to keep raising and spending.
- Tokarev says 99% of Copper's lost deals went to Anchorage, Coinbase Custody, Fireblocks, or BitGo. All four were U.S.-based, leaving Copper roughly alone on those types of deals outside the U.S.
- The real difficulty was not raising or spending but reversing the burn: headcount went from 36 in October 2021 to 236 in October 2022, slightly less than one hire per day. At that speed, culture fragments into clusters and subcultures that begin colliding.
3. The FTX post-mortem from a counterparty: collapse under growth speed, not exclusively designed fraud
- Copper signed documents to integrate FTX into ClearLoop in July 2022. Tokarev then spent months trying to get the integration moving while describing calls with Sam Bankman-Fried as hours of basketball-court conversations and “delusional stuff.” FTX staff told him Bankman-Fried had 130 direct reports; even changing a button color was not getting done.
- His conclusion is carefully hedged: FTX's conduct was “crazy” and fraudulent, but he does not think it was exclusively for conducting mass-scale fraud. He thinks Bankman-Fried lost control as the company grew from 17 to 700 people in a year.
- A problem involving leveraging or taking clients' funds through FTT and Alameda may initially have been $1M–$2M. With 134 people, Tokarev says, it could become $1B, $2B, and $5B until stopping was no longer possible. He also recalls Bankman-Fried as the first person he heard brag about having a small team, a capital-efficiency signal now common in the AI era.
4. Founder lessons: hire slowly, know your superpower, and why the CEO handoff had to be American
- Tokarev's first lesson is “hire slow.” If every problem appears solvable by adding a person, founders may be avoiding the harder work of building a scalable solution. Sometimes they need to work Saturdays and Sundays until the durable answer is clear. He is also skeptical of “we're hiring” as an automatic sign of company health.
- His second lesson is self-knowledge: “I'm a product person.” Founders should not try to make a fish climb a tree. If they are bad at hiring or administration, they should find someone who is good at that task rather than forcing themselves to cover every function.
- On the transition from Copper, Tokarev says he stepped down in December 2024 or January 2025. Friends at the headhunting firms that brought Ammar to Copper called it the cleanest transition they had seen. His rule is to back the incoming CEO and stay out of the way; transitions fail when the founder keeps blocking the new leader's plan. Jason Yanowitz adds that even Sam Altman likely has people augmenting areas where he is weaker.
- The strategy was geographic: financial institutions and any eventual exit were expected to be in the U.S., so Copper needed a U.S.-based CEO and an American center of gravity. Tokarev calls Europe “pedestrian in comparison to Wall Street” and says it is unlikely to become the center of crypto and capital markets in the next four to six years.
- Regulation was the other constraint. A product founder can build only what regulators permit; Tokarev says he has products and features sitting on a shelf that may not see the light of day for ten years, and he does not think people are capable of understanding some of them.
5. Bron/Brown: institutional MPC for individuals because seed-phrase napkins hold billions
- His wallet taxonomy has three generations. Gen one keeps the private key in memory, as with MetaMask, Trust Wallet, and Phantom. Gen two puts it on a device, as with Ledger and Trezor. Gen three is intended to solve the other nine problems, including recovery, inheritance, fraud controls, and coercion. Tokarev says institutional users have relied on MPC rather than operating private keys for roughly five or six years; he dates the technology's availability to around 2020.
- The wallet creates no single private key. One MPC share lives on the user's laptop or phone and another lives on the project's programmable server. A user can require a designated person to approve cumulative 24-hour activity above $100K or activate a 48-hour security delay above $1M.
- Hidden wallets can help with ordinary coercion because an attacker cannot tell from the interface that they exist, though Tokarev concedes they do not solve organized crime involving a gun. Lost-device recovery takes 48 hours. Inheritance can let successors regain access if the holder dies. Recovery uses the project plus a trusted third party, Crypt, which Tokarev says does quantum encryption.
- Tokarev says lost keys are creating major supply reduction as people die without telling anyone where their assets are. He cites what he thinks was Matthew Mellon's roughly $500M Ripple holding, a backup kept at a bank that burned in the Los Angeles fires, and a safe six feet beneath an office that was lost to flooding. He says tens of millions of dollars have disappeared this way.
- As crypto becomes more like digital gold and volatility declines, he argues, holders need lower self-custody risk and recoverable on-chain assets.
6. The threat landscape: weekly wrench attacks, bribed support staff, and a three-country task force
- Tokarev says physical attacks are under-discussed, with much activity in France but more globally and underreported. He estimates them at “weekly, that I'm aware of. Probably daily.” His advice is to be an unattractive target: criminals may prefer people who bought Bitcoin for $100 or who visibly expose their holdings over paranoid, well-protected public figures.
- Exchanges do not eliminate physical or social-engineering risk. Tokarev says three friends were scammed after support-center staff were bribed to leak account data. One sent $100K of Bitcoin after callers supplied transaction dates and other personal details and used a convincing British accent. He also advises checking exchange API keys; Yanowitz had found a Blockfolio-related key with trading enabled.
- Tokarev says the U.S. Secret Service, U.K. National Crime Agency, and Ontario Provincial Police formed a task force he thinks was called Operation Atlantic to fight DeFi phishing. A user can connect a wallet to a site resembling Aave or Amino and issue an allowance without revealing a private key, after which funds are drained. He compares the mechanism to the Bybit hack.
7. No equity entity: a pure utility token with KPI-based unlocks, plus Canton intents
- After reading Chris Dixon's Read Write Own while transitioning from Copper, Tokarev designed a structure with no equity entity. He says he personally has only tokens. Unlike the common model of equity for founders and VCs plus a governance token for users, this is intended to be a 100% utility token, with a subscription option for people who do not want to use it.
- Supply is fixed. Tokarev says he copied Copper's fundraising journey into a final token table rather than a cap table. Unlocks are KPI-based: more users create more unlocks, while failing to hit user targets means no unlock. He expects the first time-based unlocks, if any, to be deep in the 2030s.
- He argues that the token model produces LTV immediately because the customer is already there, unlike SaaS customer-acquisition payback that may take years. The project has about 1,000 users and $1B in assets, mostly family offices, executives, and founders. The intended word-of-mouth loop is for those users to explain the wallet to relatives over Thanksgiving, even as retail commenters ask who would launch a wallet in 2026.
- The project was the only place to buy Canton, according to the exchange. It built an intent-based cross-chain swap protocol and launched Canton pairs first because Tokarev is bullish on Canton and says “when Wall Street is writing DAML, you don't really need to know more.”
8. Stop hiring, sell door-to-door, and the crypto market read
- Tokarev says secure work with AI models through Claude Code became practical roughly three to six weeks earlier, so the team stopped hiring and remains under 40 people. He calls the current technology “the dumbest that this technology will ever be” and says it improves from morning to afternoon.
- His fundraising advice has reversed from 2021: first convince a customer to pay real dollars, then let the money follow. Brand-building still takes money, time, and patience; immediate ROI is unlikely.
- He is “super bullish on salespeople again” because corporate buyers need a person whose judgment they can trust and whose recommendation can justify a budget. They cannot fully “cover your ass with Claude.” He also expects in-person events to change but believes people will continue seeking socialization and face-to-face learning.
- Tokarev calls the moment his third Bitcoin funeral. Six weeks earlier in Dubai, a smart crypto friend told him there was now an 8% probability of a feared outcome; Tokarev replied that thinking this way signaled a market bottom. He says he bought Bitcoin for the first time in two years when Iran was attacked, after expecting the news to push it down, and Bitcoin instead rose slightly.
- His anchor is a roughly $56K realized price, which he describes as the average global Bitcoin buying price and says historically has not been breached by much. He also tracks the Fear and Greed Index, which Yanowitz says reached about 4/100.
- Tokarev is bullish on Bitcoin and Canton, owns some Zama for fully homomorphic encryption, and is considering a basket of decentralized protocols for perps, lending, trading, and options: Hyperliquid, Aster, Derive, Avax, Camino, and Jupiter.
- He calls Hyperliquid probably the most robust because of its tenacity, ferocity, and discipline. Even if its strategy continues to work and competitors find it difficult to catch up, he does not expect winner-takes-all. He compares the likely market to centralized exchanges, with Binance at about 50% of volume alongside Bybit, OKX, Bitget, and others.
Full transcript
Nothing said on Empire is a recommendation to buy or sell any investments or products. This podcast is for informational purposes only and the views expressed by anyone on the show are solely their opinions, not financial advice or necessarily the views of Blockworks. Our hosts, guests, and the Blockworks team may hold positions in the companies, funds, or projects discussed.
Alright folks, back on Empire. Took a little [music] hiatus there. Back on Empire. Very excited about this. We're recording live from DAS, so hopefully the sound's all right. Looks good. We've got the better-looking guy than me, Dmitry Tokarev. What's going on, Dmitry?
Thank you for having me, man.
Yeah. I wanted to have Dmitry on because Dmitry was actually our first sponsor ever at our first DAS in May 2019. Dmitry was the founder of Copper, which raised hundreds of millions of dollars. High valuation and big business, and now building Bron. What's going on? How's life?
Life's been good, man. I think a lot of founders can relate. We were just talking, and we roughly started businesses at the same time. When you have a smaller team, you execute faster; you're closer to clients and the product, so it's been very exciting. I think you're in it now, too—roughly, building Bron. It's been a fantastic experience, and the transition from Copper was also pretty smooth.
1. Blockworks Investor Relations
Yeah, good, man. I want to hear all about it. We've got DAS behind us, and we have 2,900 people here this year. I'd love your macro take on how the industry has evolved. We started Blockworks in December 2017. You started in January 2018.
It was the same thing in May 2019, when we were the sponsors. I remember we put together $10,000 or $20,000 or something in sponsorship fees and thought, “Okay, we've got to make it count,” because everybody was really poor. I hadn't raised hundreds of millions of dollars at that time. The theme was always the same: the institutions are coming, and so on and so forth.
We're sitting in the Franklin Templeton booth right now, so thank you, evidently, right? But I would say that a lot of people and projects that originated in crypto did so much more than they probably anticipated themselves. Some of them already went public, like Circle and BitGo, and others are gearing up to go public.
All of these shops were around 7 years ago, hoping that institutions were going to come and rescue them and save the day. But it was actually them and all of us who built this industry. The next wave people are talking about is traditional assets coming on-chain. Again, the question is: will the institutions come and save the day and bring those assets on-chain, or will players from the crypto industry do that directly and work directly with the consumer?
What do you think?
We're approaching it from both angles. On the Copper side, if institutions are there, they need infrastructure. They need companies like Copper, and there's something good that's going to happen on that front. If it's going to be working directly with the customer itself, then on the Bron side, we're ready to support.
The one thing I would say is that the way people approach investing in the digital asset space—and I would say investing generally—is changing right now. From the behavior of people, we're seeing that they're not really looking for a manager to manage their assets. They're not really looking to buy an index, whether it's an index in crypto or an index in traditional markets.
There's an increasing demand for people to understand the market dynamics themselves and then subsequently take either single-asset exposure or single-commodity exposure, whether it be palladium, gold, or silver. They want to think for themselves.
The second thing I would say is that the market cap of crypto is what? $2.5 trillion, $3 trillion?
Trillion, yeah.
On any given day—I mean, that's the market cap of Nvidia, right? All of this today is servicing a market cap of Nvidia.
Oh, sure.
No, but this is it. It's not that we're trying to hide it, I would say. If this is what it is today—servicing a market cap like this—imagine when traditional assets are going to be here and the infrastructure is going to be here, and all of those things are going to basically have to live, quote-unquote, a blockchain life.
To live a quote-unquote blockchain life, you're going to have to have all of this, essentially. All of these companies are going to have to service that industry. It's super exciting, and we definitely want to play quite a big part of it from the Bron standpoint as well.
2. The Copper Origin Story
Let's talk. I want to hear the Copper story. I want to hear the real Copper story. How much money did you guys raise?
$300 million.
$300 million?
$300 million.
At what valuation?
The latest one—I don't know if it's public. Well, you know, I'll just say: the latest one was $1.25 billion.
Wow.
Yeah. But I'll tell you: for the first couple of years, from the seed round, basically we were building a business. But in the zero-interest-rate era, you can't do that, right? You build a business, you're dipping in and out of profit. You don't really spend much money, but then nobody knows about you.
That became a problem, because when you run a custody business, you want everybody to be aware of you. Awareness creates trust. So we thought, “Okay, we've got to raise.”
You had to raise to spend money on marketing.
Money on marketing, basically. Just to be there. Just to be everywhere that's humanly possible to be. As soon as we raised, our competitors raised more.
Who are the competitors?
Fireblocks, BitGo, and Anchorage, I think. Coinbase Custody, to some extent, became a competitor.
I remember when you started, it was like you guys were an Anchorage or a BitGo, but of Europe.
Yeah. I mean, all 4 competitors—I've lost 99% of deals. Let me rephrase that: 99% of the deals that we've lost went to Anchorage, Coinbase Custody, Fireblocks, or BitGo, basically.
All 4 of these companies are in the U.S., right? So outside of the U.S., we're kind of the only one on those types of deals. Obviously, they raised, they spent, and then you have to raise and spend. Everybody did the same across the tech landscape, really. It wasn't unique to crypto, because capital was essentially cheap and people were looking to allocate more and more and more.
Then 2021 came around, and that's where the valuation insanity began. We grew from there, and there was a lot of stuff that happened in that 2021–2022 period: super-big rounds all across the board, extreme rapid growth of revenue and headcount. Then, when interest rates started increasing, you had to completely change your MO to, “Okay, well, we're now actually back to running a business,” where it actually needs to be in profit.
Yeah, exactly. Exactly.
Which was extremely hard. I think the toughest piece there wasn't so much raising capital and spending capital—it's relatively easy. Harder is cutting the burn, cutting the headcount, essentially. We went from 36 people in October 2021 to 236 people in October 2022. That's slightly less than 1 person per day who got hired, right? Then the whole system changes.
It's—I mean, everything changes, right?
The culture changes, and you can't control it. Now you basically rely on clusters of culture. Those clusters create their subcultures. The subcultures start colliding with each other. It's pretty brutal, but it's not—I mean, I know companies that have done it even faster.
Partially, I think that's what happened to FTX, because they went from 17 people to 700 in a year. I just think they collapsed under the speed of growth.
Would you do anything differently? Maybe update folks: you're out of Copper.
Yeah. I stepped down from Copper in December 2024 or January 2025, and I'm building Bron. We're building Bron with a much smaller team because, these days, different times, you know? It changes every day.
I don't know how you guys are experiencing this whole AI situation in the last 3 months, but we've been using AI since day 0. I think 3 or 4, maybe 6 weeks ago—something like that—there was a full cycle where you can now work very securely with those models. We work through Claude Code, and it changes everything, right? So we actually stopped hiring because of that. We're under 40 people.
If you started Copper over again today, would you do it differently?
Yeah, of course.
Uh, I mean, you live and learn, right? So there were a lot of lessons.
Well, if you're a founder—lots of founders listen to this—what are the lessons?
I'd say hire slow. If you think that every problem you encounter could be solved by just hiring a person, you're wrong. Sometimes it just means that you're going to have to work Saturdays and Sundays and do it yourself until you figure out a scalable solution to it that has longevity, rather than just throwing people at a problem. That's probably the main thing.
Then it's a lot to do with hiring, I would say. It's how you hire people and where you hire people. There's a lot that's being projected from the VC tech scene. For example, when people raise a round and put, “We're hiring,” I'm like, “Why? Why is this a good thing?” “We're not hiring” should be the thing.
The first company that I think changed that narrative—
Mhm. Was actually FTX. Oh, yeah?
There was a point in time when SBF talked about how small they were. He'd brag about how small they were, and I think he was actually the first person I heard brag about how small they were. Now everyone's doing it because it shows how capital-efficient you are, basically.
Yeah, yeah. Well, I mean, they went from 17 to 700 people, right?
Blew it.
At some point, because we were right in the middle, we signed documents with FTX to integrate them into ClearLoop in July 2022.
We announced it, right? There was an article and everything. Since then, I've been chasing Sam. I'm like, “Dude, where is it?” Every time you have a call, it's 4 hours of him sitting somewhere on a basketball court or something and just talking about some nonsense, delusional stuff. And I'm like, “All right, when is the date that we begin writing code to integrate ClearLoop into FTX?” He's like, “Yeah, but this is this and this is that.” Then there's philosophy, and I'm like, “Okay, this is not going anywhere.”
Then basically I talked with his team, and I'm like, “Guys, what was going on?” They're like, “Dude, you don't know what's happening here. He's got 130 direct reports right now. I'm trying to change the color of a button, and it's not happening.”
So I think the whole situation, the way it unraveled—I would say, obviously, all of this stuff that they were doing is crazy and fraudulent, et cetera. I don't think it was exclusively for the purpose of conducting a mass-scale fraud. I think he just lost control over what was happening.
There was a small problem—maybe at that time it was a small problem—of them being able to leverage or take out clients' funds through FTT leverage from Alameda's side. Maybe in the beginning it was a million or 2. Then, with 134 people, you can easily imagine that at some point, when it became $1 billion, $2 billion, $5 billion, you can't stop. You kind of go, “Oh, what's that?” Obviously, having one of your largest competitors own 10% or 20% of that, or however much that was, was a bit of an oversight, one might say.
What would you do differently at Copper, other than not hiring as many people?
We're at that stage now where I would not try to do things that I'm not good at.
You personally?
Yeah. I'm a product person. It took me a while to learn that. I build stuff. That's what I really enjoy.
It depends what kind of scale of the company you want to be running, but some people get a kick out of sales. Some people get a kick out of management. Some people have OCD and really like these admin-y tasks, et cetera. So it's really knowing yourself and not trying to make a fish climb a tree, because you can't be good at everything.
The things that you're good at—and the reason why you're sitting in the seat which you're sitting in—is that piece. It's in spite of all the other stuff that you've done in the business. If you don't know how to hire people, just find somebody who knows, and don't ask them to do other things. Just ask them to do that.
Yeah. It's really playing on each other's superpowers and asking for help in areas where you're not good at. But as a founder, you kind of expect it. You look at other founders and you're like, “Well, Sam Altman might be phenomenal at all of those things.” I guarantee not. I guarantee there's definitely a group of people augmenting something that he's not good at, and he's just doubling down on the things that he's good at.
3. How To Scale a Company
So, you're out of Copper now. Do you think non-founder-led companies can succeed at that scale? Maybe Uber's doing quite well, right? Travis is out of Uber. They've got—what's his name? Dara, isn't it?
Yeah, Dara. Yeah, yeah, yeah. He's doing a great job.
But they're already a public company, already at scale. Can you put a non-founder in a Series B, C, or D company and get big?
It's risky. I was just catching up for drinks with my friends from the headhunting firms that basically brought Ammar to Copper. He's like, “Look, from what we're seeing, this has been the cleanest transition we've ever seen, essentially.”
You bring a person in, you need to stay out of their way, right? They're going to create a culture that they think is the best. They will create a plan. You need to back the CEO that you've brought in because there's no other way. Otherwise, it's a recipe for disaster.
When things don't go well in those transitions that we keep hearing about, the majority of it is because the person never actually had a chance to implement what they really wanted to do, because there's always a founder standing in the way going, “No, you should—”
Yeah, yeah.
So it takes a bit of discipline, and it also takes a lot of self-awareness—trying to shove the ego where the sun doesn't shine—and then go, “Okay, we're here to do business. Let's focus on that.” It is weird, and it is difficult, right?
Giving up your baby, in a sense?
I never thought about this as a baby. I have 2 kids. They're my babies, you know? I love the product. I love the people I'm working with.
But you do it while thinking, “Okay, what should Copper's strategy be?” That was our thinking. One, we should be closer to financial institutions, right? Financial institutions are in America, basically. Europe—I'm sorry—is pedestrian in comparison to Wall Street. So let's call a spade a spade.
Two, what would the exit look like for the business? We thought, “Okay, it's probably going to have to be in America too, right?”
An IPO or an acquisition?
No, any exit. Any exit. It has to be West, right? You need to have a US CEO—an American CEO, a US-based CEO—because otherwise, such a core, critical piece of infrastructure just needs to have an American center of gravity. That's what we wanted to achieve with this move, essentially.
When you give up control, you basically know the reasons why you're doing this, because unless Europe or the UK turns around and says, “Okay, we're now going to be the capital of not only crypto but the entire capital markets,” it's not going to happen in the next 4, 5, 6 years. I think that's the main motivation and the main reason.
The other reason is that when you work in a regulated business, the limits of what you can create as a product guy are basically bounded by what regulators will permit you to create. I built stuff that's sitting on the shelf there. That's never going to see the light of day over the next 10 years, and now I'm even more certain of that than I was.
Mhm.
No, I thought it was just going to take them a while to understand it. I think certain things that I've built and shelved—I don't think people are capable of understanding them.
Understand what?
The features, the pieces of the products that we've built, essentially. Because that's what I would do if I were running Copper: I would be building stuff. In terms of building stuff, if you're limited to what regulators permit you to build, there's only so many things you can build.
4. ZKsync Ad
ZK Sync is the bank stack of Ethereum. ZK Sync's Providium is the only Ethereum secured platform that is purpose-built for institutions [music] that demand privacy, compliance, and full control of their data. ZK Sync enables financial institutions to thrive in the digital [music] assets economy by launching and monetizing their own chains, tokenize real-world assets, build programmable markets, and settle cross-border in [music] real time. Visit zksync.io to learn more about the bank stack of Ethereum and book a demo. As always, investments in blockchain technology involve risk. Terms and conditions apply. Hey all, Blockworks co-founder Michael Ippolito here. Quick break to talk about something we've just launched, Blockworks Investor Relations. As the market shifts toward institutional capital, investors want more transparency, more standardization, and a higher level of professionalism. But, the traditional IR model is slow, manual, and not built for how crypto works. If you're building on chain, your data's already live, your business is already transparent. The challenge is turning that into a clear, credible story for investors. That's exactly what we're solving with Blockworks IR. It's a single platform that brings together real-time analytics, branded investor portals, and hands-on advisory support so you can communicate what matters. If you're an on-chain business looking to level up your investor strategy, check out Blockworks Investor Relations at blockworks.com/investor-relations. All right, back to the episode.
What are you doing with Brown?
There was a whole gap that we identified, which was actually shocking to find out. Six or 7 years ago, at one of your events, I remember explicitly that 2 founders of crypto custody companies were having almost a fistfight over whether it would be MPC or HSMs.
Yeah, yeah. I remember. I don’t know who that was.
Exactly. That was the last battle, right? Since then, everyone’s using MPC.
MPC is essentially technology where you don’t have private keys. All of you probably have to write down seed phrases and create private keys, but institutions just haven’t done that for the last 5 or 6 years. No one sitting at BlackRock—or rather, Coinbase servicing BlackRock—is writing a seed phrase for BitGo Vault 1.
Right. That doesn’t exist. We lost the key.
It’s funny, too, but this is how the crypto industry basically functions today when it comes to retail. We’re talking about retail with billions of dollars sitting on napkins with 24 words.
There are 3 generations of wallets. There’s gen 1, which appeared straight out of the gate: it generates your private key, and you write down the seed phrase, like with a software wallet. They still exist because they’re quite convenient. Your private key just lives there in memory, like with MetaMask, Trust Wallet, and Phantom.
Then there’s gen 2. People think, “If this laptop gets hacked and the memory is drained, and there are North Koreans sitting there sniffing everything, it’s easy to steal the private key. Let’s put it on the device.”
Ledger, Trezor, and so on. Let’s put it on the device.
That solved that problem. Then gen 3 solves the other 9 problems that remain. That’s basically what Brown is doing. We’re a gen-three self-custody wallet.
What are those problems?
If you’ve lost everything you’ve ever created on Brown—your devices and so on—we can help you recover. You can’t lose that wallet. If you get hit by a bus, you can appoint successors, and your family and relatives are going to get access to your crypto.
This is probably creating the most supply reduction over the next 10 years. It’s not the protocol itself; it’s people dying and not telling anyone where their stuff is. That’s already happening with Matthew Mellon, I think—the guy who had a bag of Ripple worth about half a billion dollars. That’s gone forever.
There’s another increasing problem that people experience. You need to protect against fraud. You can’t give your lawyer or accountant your seed phrase; that’s crazy. You need to solve for that, and we’ve solved for that.
Weirdly, a lot of that is happening in France, but so much more is happening globally and is underreported: physical attacks.
Physical attacks are very under-discussed, right? It’s insane. I’m part of several groups, and we can turn this into a very sad podcast. I’m of 2 minds about whether to tell people what’s actually going on.
Give me the scoop.
If you have anything anywhere on a wallet with a seed phrase, install Brown and don’t use that wallet. That’s my advice. If somebody rocks up to you and politely asks you to give it up, you’re going to have to give it up, right?
But how frequently are these attacks happening?
Weekly, that I’m aware of. Probably daily.
Is it mostly big public figures, or is it people they identify through something like the Ledger leak?
It’s mostly big public figures. When you talk about physical attacks, the main thing is to make yourself extremely unattractive. You don’t want to be an attractive target.
If you’re a public figure, the risk associated with robbing you is pretty high because people will care. That’s one thing. The second is that you’re potentially thinking about those things and have security in place.
So they’re not really going after public figures?
Yeah, because who are you going to go after—Michael Saylor? That’s crazy. Who do you want to go after? Criminals figure out who the people are who bought Bitcoin for $100.
And they’re 80 years old, they live in Elbe [?], and they don’t have any—
Exactly. Or somebody who’s flashing their portfolio, for example. You can tell from someone’s demeanor who’s serious about their security, who’s paranoid, and who’s not. You don’t want to rob paranoid people because they have processes in place.
What’s the best way to protect your brand-new motorcycle? Park it next to a brand-new Ducati. People are going to try to steal the Ducati first before they get to your bike.
There’s a lot of that happening, and we’re not going to move forward as a space with all of these problems. I’m surprised we actually got to this state with the infrastructure that normal people have.
Yeah, but most people don’t do that. Most people just use Coinbase or Kraken. I mean, most people use centralized exchanges. They all went to the exchanges.
If you’ve got access to those exchanges, you can lose access to those funds during a polite visit by somebody with a machete.
The biggest problem there is actually social-engineering scams. That’s another way to lose your crypto.
Yeah, there’s the text message saying it’s from Coinbase.
The most popular method lately has been bribing a support-center person to leak account data. I had 3 friends who got scammed that way, and these are not stupid people.
One of them was made to transfer $100,000 worth of Bitcoin to a different address. I asked him—because he knows me, and I’m much less reserved when I talk about these things off camera—“This is why we're building Brown. Be paranoid, install everything,” and so on. I asked him, “Did you think for at least 10 seconds during this whole episode that they might be scamming you?”
He said, “Yeah, but they were so convincing. They had a fantastic British accent, and they had all my details.”
So they must have bribed a customer-support person.
They bribed a customer-support person. They basically told him which dates he sent which transactions.
Wow.
Those are the kinds of details they had. If somebody calls you from Amex, they might say, “Let us identify ourselves. On this day, this is the amount that was charged to your card. This is when you created your card, and this is the address where we shipped your card.”
You’re like, “Okay, you can’t really ask for more in terms of identification,” right? It’s the same thing with crypto. People willingly give up their crypto because someone tells them, “Your account isn’t safe. We just need a confirmation code and number.”
Sometimes that’s rarer, but another way you can lose your crypto is by simply generating an API key. People should look at the API keys generated on their accounts, because sometimes, even if there’s whitelisting, there’s a way to steal through the exchange.
What? I just saw an API from Blockfolio that was feeding into Blockfolio and had trading enabled.
Oh, yeah.
No, no, it never—no one took advantage. Yeah, I just found it the other day.
Yeah, so that’s it. The other thing these days is that, if you’re using DeFi services, it’s gotten so bad that the U.S. Secret Service, the U.K. National Crime Agency, and the Ontario Provincial Police formed a task force called Operation Atlantic, I think, which is basically fighting DeFi phishing scams.
You can hook up your wallet and issue an allowance on some website. No private key is revealed, nothing—your funds just get drained straight away. It’s similar to the Bybit hack, actually. You issue an allowance, and that’s all you have to do. It looks like Aave; it looks like Amino.
So why do you guys solve this when someone like Coinbase can’t solve it?
On the Coinbase side, there are a few things—
Or maybe don’t just compare it. On the DeFi phishing side of things?
There are a few things. Bronze doesn’t solve it, so here’s the thing: when we create a wallet for you, there’s no private key that’s created. That’s how institutions do it. You know it, right? That’s basically been the default case for the last 5 years. There are no private keys. Nobody in the institutional space operates private keys, right?
It's a seed phrase wallet. People use MPC, which basically creates key shares, or shards. One share lives on your device, whether it’s a laptop or a mobile phone, and one share lives on a Bronze server, which you program.
You program the server—the share—to do things. For example, you can say, “If my cumulative transaction volume over the last 24 hours exceeds $100,000, I want this person that I added to my account to approve my transaction. Otherwise, don’t sign with me.” Or, for example, “If we exceed $1 million, I want a 48-hour security delay activated on my account.”
Then we have protections against random crime. If somebody on the street asks you to open your wallet, you can basically hide wallets, and from the interface you won’t be able to tell whether they even exist.
That’s my favorite.
Yeah, it’s pretty cool, right? Obviously, it doesn’t help with organized crime because, if people point a gun to your head, you’ll quickly remember all your hidden accounts. But the policy engine does help with that.
If you get hit by a bus, there’s a full recovery mode that could be enabled by your successors. That’s your inheritance feature. People lose devices every single day. They just go through a recovery process, which takes 48 hours, and regain access to their lost key share.
It’s essentially Bronze and a trusted third party appointed on your behalf as a backup.
Yeah, Bronze and that trusted third party—is it like a sibling or a friend or something?
No, no. We work with a company called Crypt. They do quantum encryption. You inform them that you would like to recover your share, and after the security delay, we essentially print back the key.
A three-party multisig?
Yeah, yeah, yeah, yeah, yeah.
Interesting. Why are you going after individuals? Isn’t it a worse business model?
Look, I think in order for us to advance as a space, this needs to be solved. We can’t be operating in a space where people are going to lose their 401(k) savings because they forgot where they stored a napkin, or because of some other mistake. I mean, the stories I’ve heard—
One guy had a backup at a bank. The bank burned down in the L.A. fires, right? And the bank was like, “We don’t know where it is.” Gone. Another guy basically had it in a safe 6 feet under his office, essentially in a basement. The floods came, and it was gone.
I’m talking about tens of millions of dollars gone, right? We can’t continue as a space like that. It might have been acceptable when the volatility was so insane that one day you had $1 million, another day you had $100 million, and another day you had $50 million.
But now that it’s dying down more and more—and it will turn into more of a digital gold-type scenario—you would want to have a reduced level of risk associated with self-custody of such assets. It’s extremely important.
Crypto is part of the story of assets that live on the blockchain. We will see other assets moving on-chain, and if you want to hold those assets, you should be able to recover them. You should be able not to give them up that easily.
Is there some new technology that made this possible for you guys?
MPC, essentially. We just took what works in the institutional world. We took what we’ve basically been doing in institutions and applied it here.
And competitors have been using it since 2021?
Yeah, it’s basically been around since 2020.
Exactly. Since 2021, everybody’s been using MPC.
Exactly. Since 2021, everybody’s been using MPC. We just took that and applied it. Why? Because people deal with $20 million, and that’s a lot of money. But $2 million is a lot of money, and $20,000 is a lot of money. It should be accessible to everybody.
5. Why Bron Is Launching a Token
We’re democratizing access to technology that’s been used by institutions for the better part of half a decade.
How’s it going so far? Did you raise money?
Yes. We did something differently. I read a book by Chris Dixon from Andreessen called Read Write Own.
As I was transitioning from Copper, I sent you pictures from the beach. That’s where I was reading the book because I wanted to brush up on what this whole Web3 thing was, right? I didn’t learn much that was new, but it succinctly explained Web3 and gave me an opportunity to explain it to people.
Web3 allows your clients—the people who help you early on in building a business—to grow with you.
Oh, this is where the Bronze token comes in.
Yeah, and then we built the tokenomics, essentially. Here’s what we’ve done differently: we don’t have a company with equity. It doesn’t exist. There’s no equity. There is no entity that has equity.
I know it’s been popular in the blockchain space to have an equity company for founders and VCs, and then a token for governance.
Yeah, great. Now we’re changing that. Obviously, that was the wrong model. I mean, for whom?
For the user.
For the user, exactly. For the users, it sucks.
So we decided it’s just going to be a token. There’s an opportunity to pay a subscription if you don’t want to deal with the token. We’re not shilling the token; it’s an option. It’s a pure, 100% utility token.
There’s no equity in the company. It doesn’t exist. I only have tokens.
Is it a fixed supply or unlimited?
It’s a fixed supply, but here’s another part. What people normally do is time-based unlocks. What we do is KPI-based unlocks. The more users we have, the more unlocks we have. If we don’t reach the user targets, we don’t unlock.
I would think our first time-based unlocks would be somewhere deep in the 2030s.
Why would you have a fixed supply? That model never made sense to me because it’s taken from Bitcoin, right? That’s why everyone does a fixed supply. But Copper didn’t have a fixed supply of equity. You go raise another—
Yeah, totally. It’s actually a good question. I took Copper’s fundraising journey and said, “Okay, before exit, there might be another round. If we do it this way, this is how it would look.”
We literally copy-pasted it and said, “This is the final state of the token table.” It’s not a cap table; it’s a token table. That was the idea.
You’re right—there are definitely ways of doing it where you dilute over time, raise more, and so on. But times change. You don’t optimize the business for infinite fundraising cycles. You build something sustainable that grows in line with the user base you have.
What’s good about this model is that, normally, in a SaaS world, you have customer acquisition cost and lifetime value. You can calculate your LTV over a 5-year period and figure out how many years or months it will take for the customer to pay back the cost of acquiring that customer.
The good thing about the token model is that you get your LTV straight away. You don’t have to wait 5 years to get the LTV because the customer is already there.
Again, it’s all super experimental. It’s 100% utility, but so far it’s been working very well. We’re going to continue improving how we tell the story around it.
Some people say, “Dmitry, I’m not too sure about your token.” No problem. You can just pay a subscription.
How were you guys the only place to buy Canton?
We were. We were. We were.
Canton’s been a great project, and I’m super bullish on it because when Wall Street is writing DAML, you don’t really need to know more.
6. Building Bron: How To Fix Crypto Custody
People’s roots are already going very, very deep on Wall Street. If there are DAML engineers sitting at banks and writing code, right? What we’ve done is look at how the swaps ecosystem evolved and what people are doing in terms of swapping assets—Uniswap, 1inch, and so on. We stumbled upon an intent-based approach, an intent-based protocol, and we really liked it. We thought, “Why don’t we launch intents?” So we launched intents.
It allows you to do cross-chain swaps. The first pairs launched on intents were Canton pairs.
Because you were bullish on Canton?
Well, we added Canton because we know the team. I worked with the team during the Copper days, too. As Brown went into production, it was like, “Okay, what should we focus on?” We were chatting with the team: “What do we launch on Canton?” So, we added Canton.
With pretty much any blockchain we add, we automatically add the ability to do swaps. So, you can do that, and it’s extremely capital-efficient.
How do you track success? Is it AUM? Is it the assets on the platform?
Yeah, we’ve got AUM, but it’s a proxy. We’re yet to determine what our North Star metric is, but it’s the number of users and, in the end, the average amounts that these guys are holding on Brown.
We’re at about 1,000 users at the moment. We’re at about $1 billion of assets.
Wow.
Mostly family offices, executives, and founders. The biggest problem—this is going to be the exercise for the next 9 months—is that one influencer who does the Brown wallet review posted it on YouTube. The comments were, “Who in their sane mind is launching a wallet in 2026? What’s the matter with these people?”
Obviously, retail has no clue who the Copper people are, right? The problem that I’m solving is that the listeners of your podcast probably don’t know what Brown is, or are learning what Copper is. Some do, but the guests on your podcast are Brown users.
Yeah, and that’s exactly it. I mean, that’s why I want to have you on. We usually don’t bring on early-stage people.
It’s all the founders and CEOs. I’m like, “If only…” But again, it’s kind of weird to go and ask them, “Hey, can you tell the whole world where all your stuff is?”
Yeah, and then they’re like, “Dmitry, what’s the matter with you? Just leave me alone.”
Yeah, no one wants to say where their—“Stop texting me.”
It will come with time. Bridging this gap and explaining it to people is important. Word of mouth is pretty strong as well, because probably in your family circle, and definitely in my family circle, when people have a crypto question, they come to me. It’s like, “Dmitry, what do you do with this?”
That was always the objective: to bring in those cool people who are technically savvy, founders of crypto projects, or executives working in digital assets or on Wall Street. Get them as users, so then on Thanksgiving Day, while carving the turkey, they’ll tell people how to set it up, right?
Yeah, exactly.
What did you learn about raising money, and do you think you raised too much money?
Nobody gave you the option to raise less money in 2021 or 2022, because if you raised less money, you didn’t exist. It was as simple as that.
We had a commercial with Rebecca Ferguson touching the waterfall, and there was a 60-person crew that flew to Iceland to film all of this stuff. It was insane, but you had to do it. It cost so much money that I did not watch Dune: Part Two so as not to remind myself how much we spent on that commercial.
[laughter]
So, I’m like—
On that commercial? A lot, man. A lot. Ask Tyler.
Yeah, I will ask Tyler.
[laughter]
7. Lessons Learned From Fundraising
I think there might be some people working on stuff like that who are involved in revealing that side of things. But nobody gave you an option. There’s no option right now with AI stuff, because it seems like you have no option but to raise massive amounts of money if you’re an AI project.
Yeah. Man, AI—Claude has been ruining my life since February. I’ve been deploying stuff. I’m techy, but I’m not normally the person who would set up instances on AWS with S3 buckets and cron jobs.
It can do everything. Exactly, exactly.
So, I’m like, “Wow, this is so cool.” I wish I’d had that at 24 as an entrepreneur. It would have been so amazing.
Right now, if you’re a founder, if you’re 24 and weird, and you just want to build stuff, there’s never been a better time. Somebody said recently, “This is the dumbest that this technology will ever be,” right? At the moment.
You feel it. Every month it gets better.
Dude, every morning to afternoon, you know?
Yeah.
In terms of raising more capital, I’m not sure. I don’t think it’s a good thing at this day and age. The most important thing is a customer, right? If you raised a lot of capital, it’s easier to convince people to give you money than to convince a customer to pay you money, right?
Yeah, yeah.
The very first thing you should do—and again, it’s just not to waste your time—is convince a customer to pay you real dollars for the product that you’re building or have built. Then the money will follow. I would not start with a fundraise.
Yeah. What did you learn about spending a lot of money on marketing? Other than spending with Blockworks, which is a really, really good idea.
Yeah, of course. One thing I’ve learned is that building a brand costs money. There’s no magic secret where you spend $2 million here and then, all of a sudden, you’ve got a brand.
It takes time, but it’s totally possible. If you’re present and creating value for people, you will build a brand. Patience pays off as well. Nobody knows you, nobody knows you, nobody knows you, and all of a sudden, you’re a household name.
How’s that?
That’s how I feel about Blockworks. Nobody knew us for even 5 or 6 years.
Right. And now it’s crazy.
Exactly. A lot of people are like, “I’ve loved Blockworks since day one.” And I’m like, “No, no, no. It was like, ‘Why did we spend $20,000 on this?’” That’s the type of conversation people were having.
Yeah, totally. Let’s give it a go. Let’s try it, right?
Brand-building on the marketing side takes time. You’re not going to see ROI straight away, and it will look like the dollars you burn.
One thing with this new AI situation, which I think is—let me put it this way—I’m super bullish on salespeople again.
Really?
Yeah. Fast-forward 2 or 3 years, with all the AI, with people going to stop buying like that, people are going to go back. I think we’re going to go back to something almost closer to door-to-door salespeople, because this is going to be such an underappreciated life hack.
Do you think you’ll just go into Claude and be like, “Hey, Claude, what’s the best CRM?”
I don’t know. CRMs are not going to—It’s going to be difficult to continue CRM businesses as they are right now. Where do the salespeople go?
Salespeople are—well, it depends on what kind of buyer you are, right? If you’re buying for a corporation, you can’t point to Claude. Why does McKinsey exist? You pay them to cover your ass.
You won’t be able to cover your ass with Claude. You’re going to jump on a call with Jason or Dmitry, who’s going to be like, “Trust me, this CRM system is going to work,” so that you commit your budget. Then your CTO isn’t going to fire you because you just wasted $2 million on this project, right?
You need a person to tell you that, so you can look them in the eyes and see how convincing those eyes are. I’m super bullish on that.
I think IRL events are also going to change. The whole COVID era was not really nice, and from the Copper standpoint, it bridged us in terms of capital. All of a sudden, it was a level playing field because it was booth-to-booth sales, and you couldn’t go to events, right? We didn’t have the capital to be at events.
That was great, but that was the past. People want to go and talk to people.
People want to hang out together and learn from each other. We’re already way dumber than AI at the moment, so we’re going to have to live with our monkey brain, which requires socialization and a couple of beers.
8. What To Own in 2026?
A couple of beers, exactly. This is great, man. Anything else that you’re thinking about these days?
Well, now that you mentioned it, I’m thinking about a couple of beers.
Yeah, there you go.
Well, look, it’s interesting how the space is going to move. I would say that this is the 3rd Bitcoin funeral I’m at. Last year, we were waiting for it. We knew that this bull market was at some point going to end, and then everybody was going to be like, “Well, told you.” And then there were quantum computers that were going to basically make all of this disappear.
It slightly started to go away, even compared with 6 weeks ago, because 6 weeks ago I was in Dubai with a friend of mine who is a very smart crypto guy. He said, “Listen, I’m worried. There’s now an 8% probability.” And I’m like, “Dude, you realize that this is the bottom of the market? If you were thinking that way, this is the bottom.”
I bought Bitcoin for the first time in 2 years when we attacked Iran, and Bitcoin went up a little.
Yeah.
I thought it should go down because it goes down on terrorist attacks. It kind of has been going down on any bad news.
The realized price right now is about $56,000. So, realized price is basically—look, what’s the average global buying price for Bitcoin?
Oh, across everybody?
Yeah, across everybody, right? And people are just really bad at facing paper losses. Historically, it did not dip anywhere near as low as their realized price. It’s a pretty solid metric. If you look historically, it’s quite interesting.
I’m also tracking the Fear and Greed Index. That’s been quite interesting to watch.
Yeah, it got to, like, 4 out of 100.
Yeah.
Then what do you buy? You buy Bitcoin? Ignore your own token.
You buy Bitcoin. I’m pretty bullish, and, you know, not financial advice or whatever disclaimers I need to make.
Yeah, we’ve got it.
Yeah, but you have it on your website, right? Everything?
We have all the nondisclosures.
So I have the big 3, let’s say. Bitcoin is all. I’m super bullish on Canton because that’s something which I believe Wall Street is integrating, and not just talking about integrating. I just think it’s going to take a little bit of time before those things ship to production and then actually reach proper bank-like, industry-like scale.
The other things which I’m thinking about—so, I got a little bit of Zama.
What is Zama?
Fully homomorphic encryption.
Yeah.
It’s basically on existing rails. On Ethereum, you can just make your USDT disappear and move it around, et cetera. So I think that’s pretty powerful.
At the moment, I’m thinking of potentially accumulating a bag of decentralized protocols for perps, lending, trading, and options, essentially, because I think there are quite a few things that we are working on in that regard. So, a little bit of a teaser.
The decentralized venues—like Hyperliquid?
Like Hyperliquid, Aster, Derive, Avax, Camino, Jupiter—there are quite a few protocols. Some of them, obviously, I think are going to become absolutely massive.
Hyperliquid is already showing what’s possible. These guys are not stopping. They have 3 markets now; they have 4 markets. Who do you think becomes massive there?
At the moment, the most robust is probably Hyperliquid, of course, with the tenacity, ferocity, and discipline that these guys have built. If that continues, if their strategy works—and it looks like it’s pretty solid—it’s definitely going to be hard to catch up, but I don’t think it’s winner-takes-all.
Yeah. It’s going to be the same situation we’ve seen in centralized exchanges. There’s Binance with 50% of all the volume, but there’s Bybit, there’s OKX, and there are Bitget and all the others, which are also seeing their share of the volume.
So that’s basically a trend that I’m super interested in keeping a close track of.
Nice. Awesome, man. Good to see you.
Likewise. Thank you.
Yeah, thanks for having me. When are you going to sponsor?
We’re going to discuss this now over those 2 beers that you mentioned, probably.
All right folks. Good to be back on Empire. Missed you guys the last 2 months. And uh Dmitry, congrats, man.
Thank you so much.