Hope Sterling: Can I just start with a confession? I thought I understood crypto — like, reasonably, for a normal person — and then I hit this one statistic and realized I do not understand humans at all.
Hope Sterling: Fifty-nine percent of crypto users say they want self-custody — they want to hold their own private keys, no exchange, no middleman. And then only seven-point-five percent do it. That's the gap. More than half of people literally want a thing and then don't take it, and I cannot figure out why.
Juniper Vale: I'd say 'I don't buy that it's confusion' — because it's not. Those people know what self-custody is. They're opting out for a reason.
Hope Sterling: Fear of losing the key? Like — what even happens if you lose it?
Juniper Vale: The assets are gone. No appeal, no recovery. The blockchain doesn't know who you are — it only verifies the private key that signed the transaction. Whoever has the key has the money. So yeah — the fear is earned.
Hope Sterling: But then — okay, so the alternative is parking everything on an exchange, and then FTX happens. Users in insolvency, competing claims, potentially zero recovery. Bitcoin was literally created so that couldn't happen to people. Like, the whole point of Bitcoin was 'be your own bank, trust no intermediary.'
Juniper Vale: So you're stuck between two real dangers, and the question — the one I don't think has an obvious answer — is what does it actually mean to hold crypto, and why is holding it yourself so costly that most people keep choosing the version that already burned them?
Hope Sterling: Wait, but — hold on — when you say 'hold the key,' what does that actually mean? Like, is my wallet app on my phone... is that the key?
Juniper Vale: No — and this is the click. The app is just a window. Think of it this way: imagine your house key didn't just open your door, it *was* your legal title to the house. Whoever physically holds it owns the house. No deed, no court, no identity check. That's a private key. The blockchain doesn't know your name — it sees a signature, confirms the key produced it, and that's it. Transaction authorized.
Hope Sterling: So the wallet app is... what, like a picture frame around the key?
Juniper Vale: Pretty much. It's a display. The key lives underneath — and wherever that key actually exists, that's who owns the Bitcoin. Full stop.
Hope Sterling: Okay that is — I mean that reframes everything because I always thought the exchange was like, holding my account. But they're holding the actual key.
Juniper Vale: Exactly. And — okay, this gets uncomfortable. Because the backup for that key, the thing that lets you reconstruct it if your device breaks, is a seed phrase. Twelve to twenty-four words. Just words, like a weird poem. And if those words are gone — I mean, if someone throws away the notebook, or the house floods — the assets are permanently unreachable. Not frozen. Not recoverable via customer service. Gone. The math doesn't care.
Hope Sterling: No customer service line. Like, literally none.
Juniper Vale: Picture someone's grandmother, right — she has a hardware wallet her kid set up for her. She writes the seed phrase on a notecard and slides it into an old recipe box because that's her safe place. The card gets tossed in a kitchen cleanout. Those assets are as gone as if they never existed. Nobody can override that. That's what asymmetric cryptography means when it becomes your bank.
Hope Sterling: And the recipe box thing — like, that's not even negligence, that's just being a person. But wait, okay, so — that's self-custody failing. What about the other direction? Because I keep thinking about the people who looked at all that and said 'too risky, I'll leave it on the exchange,' and then FTX collapsed.
Juniper Vale: And those people didn't lose because of a bad password. They lost because they never held the key. The exchange held it.
Hope Sterling: Which means — wait — they were inside the insolvency? Like, legally their Bitcoin was just... disputed property?
Juniper Vale: Competing claims. Courts don't have clear jurisdiction on who owns custodied crypto when the custodian goes under — it's genuinely contested across legal systems. You might be a creditor. You might get cents on the dollar. You might get nothing. And this isn't a FTX-only problem — Dan Wasyluk and his colleagues lost 750 Bitcoin when Moolah collapsed. An actual person, actual coins, gone. That was an early case, and the pattern just keeps repeating.
Hope Sterling: Seven hundred and fifty Bitcoin. I mean — at any recent price that's — that's not a rounding error.
Juniper Vale: And the mechanism is the same every time. Delegated custody concentrates all the risk in one place — one target, one set of keys, one insolvency filing. The attack surface for the exchange is every customer's assets simultaneously.
Hope Sterling: Okay and here's the part that actually makes me angry — because when you're on an exchange, you also just... gave up censorship resistance entirely, right? Like a regulator can call them Monday morning and freeze everything before you even wake up.
Juniper Vale: The custodian holds the keys, so yes — a government, a regulator, an insolvent trustee, whoever — they can instruct the custodian to freeze or seize. You have no recourse because you have no key. The censorship resistance that Bitcoin was built to provide disappears completely the moment you delegate.
Hope Sterling: Which makes me wonder — and I don't want to get ahead of ourselves — but there are these newer approaches, like MPC wallets, threshold signatures, things like Zengo, where they're trying to split the difference. Because what we just described can't be the end of the story.
Juniper Vale: Right — but the part that doesn't fit the clean story is that MPC isn't magic. Zengo, for example — it splits the private key material between your device and their servers, so neither half alone can move funds. Biometric recovery built in. No seed phrase to lose in a recipe box. That's real. That's genuinely better.
Hope Sterling: Okay wait — so it's like, isn't that just... a fancier exchange? Like, Zengo's servers are still in the picture.
Juniper Vale: That's exactly the right push. You've moved the trust — you haven't removed it. Zengo's infrastructure goes down, gets compromised, or — I mean, what if they have an insolvency event? The trust problem got relocated to a layer that's actually less visible than the exchange was.
Hope Sterling: More opaque. Which is somehow worse?
Juniper Vale: Potentially, yeah. At least with an exchange you know you're trusting them. With MPC the user feels like they have self-custody, but there's a server somewhere that is load-bearing for every transaction they make.
Hope Sterling: Okay and threshold signatures — institutions use those, right? Like, multiple parties have to sign off before anything moves?
Juniper Vale: Same principle, bigger table. You need a subset of, say, five keyholders to co-sign — no single person can run off with the funds. Distributes the attack surface. And regulators — the CFTC, the SEC — they're trying to build auditable baselines around exactly this. The CMTA put out a Digital Assets Custody Standard to give institutions a framework for assessment. But — and this is what I want to be honest about — those frameworks acknowledge the tradeoff. They don't dissolve it.
Hope Sterling: They're just like... making the tradeoff legible? Putting a label on the thing that still exists?
Juniper Vale: Think of it as a spectrum. Every single solution — full self-custody, hardware wallet, MPC, exchange, institutional threshold sig — it's just a coordinate. Move toward control and you inherit operational risk. Move toward recoverability and you inherit counterparty risk. Nobody has found a point on that line that gives you full censorship resistance and full recovery simultaneously. That's not a failure of engineering. That's the structure of the problem.
Hope Sterling: That's the part I keep sitting with — like, it's not a bug. The reason MPC moves the trust instead of removing it, the reason a hardware wallet still requires you to not throw away a notecard — that's not because engineers weren't smart enough. That's just... asymmetric cryptography. The math doesn't have a compassionate override.
Juniper Vale: And that's actually the part that matters most to me about this whole thing. Because if exercising self-custody safely requires either serious technical depth or enough money to hire someone who has it — then the system that was supposed to distribute financial control might actually just concentrate it among people who can afford to do it right.
Hope Sterling: Which is the opposite of the Bitcoin promise.
Juniper Vale: Yeah. Quietly, it is.
Hope Sterling: I don't think I have an answer to that. I just — I think I needed to actually see the shape of it. That it's structural, not a thing anyone forgot to fix.
Juniper Vale: That's an honest place to land. Thank you for thinking through it with me — genuinely.
