Short answer. The seed phrase does not disappear — the single copy you have to guard does. Modern wallets split the key material so that no one party, including the company, ever holds a usable whole. You sign in with email; the key is assembled only to sign, only on your instruction. What you give up is described further down, because every page like this leaves that part out.
Twelve or twenty-four words that encode one private key. Whoever has them controls the money — completely, permanently, with no appeal. That design has one great property and one terrible one:
Most crypto losses among ordinary people are not exotic hacks. They are those three sentences.
Instead of one key in one place, the key material is split into shares held by different parties — your device, secure infrastructure, sometimes a recovery share. A signature can be produced when the required shares cooperate. No single holder can sign alone, and no single holder can reconstruct the whole key.
You will see two families of this:
| Approach | What it means |
|---|---|
| MPC (multi-party computation) | The key exists only as shares. Signing is a joint computation between them; the full key is never assembled anywhere |
| TEE (trusted execution environment) | Key operations happen inside isolated hardware that even the machine's operator cannot read into |
The practical result is the same: there is no moment where a company holds something it could use to take your money, and no twelve words for you to lose. hho.cash uses this model through Privy, which secures key operations in a TEE.
Yes — and the reason is worth understanding, because "log in with email" is exactly what a custodial exchange also offers, and the two are not the same thing.
On a custodial exchange, the company owns the keys and your balance is an entry in their database. They can freeze it, and if they fail you are a creditor. With split-key self-custody, your email is how you authenticate to your own key shares. The company cannot sign without you, because it does not hold what it would need.
The test that matters: can the company move your money without your instruction? On an exchange, yes. Here, no. Everything else — the login screen, the recovery flow — is interface.
Here is the same question answered specifically about hho.cash, in the form nobody publishes:
| Can hho.cash… | Why | |
|---|---|---|
| Move your money on its own | No | We cannot produce a signature without you |
| Freeze your balance | No | Nothing to freeze — it is not in our custody |
| See your seed phrase | No | There is not one to see |
| Lose your money if we shut down | No | The wallet is yours; the address exists independently of us |
| Send within limits you explicitly allowed | Yes | See below — this one is real and worth reading |
To make sending free and instant, we ask you to grant a limited permission: our backend may submit transfers on your behalf, inside caps enforced cryptographically — a maximum per transfer, a maximum per day, and an allow-list of what it can interact with. That is what makes a send feel like a message instead of a gas-fee ritual.
It is a real delegation, and we would rather write it here than let you discover it later. It is bounded, it is revocable, and it does not extend to your whole balance. If you would rather sign every transaction yourself, that path stays available.
This is the question the whole design turns on, and the honest answer has two halves.
The wallet is not stored on the device. Sign in on a new phone and your balance is there. This is the case that used to destroy people, and it no longer does.
Your email is now the credential that matters. If you permanently lose access to it — not forgetting a password, but losing the account itself — recovery becomes difficult and may be impossible.
So the practical advice is unglamorous: put a strong, unique password and two-factor authentication on the email address you sign in with. You moved the risk from twelve words on paper to one email account. That is a much better place for it — but only if you treat it that way.
Seedless wallets are not strictly better. They are better for a particular person, and here is who they are not for:
Seedless wallets are built for money you use: spending, sending, saving amounts that matter but are not your entire net worth. That is the trade — and it is a good one for most people, most of the time.
A self-custodial wallet with no recovery phrase for you to write down. The key material is split into shares held by different parties, and a signature requires them to cooperate, so no single party — including the company — can move your funds alone.
It removes the most common ways people lose crypto: losing the phrase, photographing it, or being tricked into typing it somewhere. It adds a dependency on the provider's infrastructure and makes your email the credential that matters. For everyday amounts that is a good trade; for very large long-term holdings, hardware cold storage remains the conservative choice.
No. Email is how you authenticate to your own key shares, not how a company holds your funds. The test is whether they can move your money without your instruction — on a custodial exchange they can, with split-key self-custody they cannot.
Nothing is lost. The wallet is not stored on the device — sign in on a new phone and your balance is there.
That is the serious case. Your email is the credential the account depends on, so permanently losing the account itself can make recovery difficult or impossible. Use a strong unique password and two-factor authentication on it.
Your wallet address and its balance exist on the blockchain independently of any company. Reputable providers publish an exit path so keys remain accessible. This is nonetheless a real dependency that a paper backup does not have.