Securing a crypto exchange account: 2FA, passkeys, and the SIM-swap problem
Financial disclaimer, not financial advice. This article is general information and education only. It does not recommend buying, selling, or holding any crypto asset, fund, or product, and it is not personalised financial, tax, or legal advice. Crypto assets are volatile and speculative and can lose all of their value. Consider your own circumstances and, where appropriate, speak with a qualified professional in your jurisdiction before making any financial decision.
Most exchange accounts are protected by whatever the sign-up screen offered first: a six-digit code by text message. That is better than a password alone, and it is the weakest two-factor method in common use, for two separate reasons that are usually collapsed into one warning. The first is the SIM swap: an attacker who moves your phone number receives every code your accounts send you. The second is quieter and now larger: a fake login page, relaying in real time, can use a typed code, from SMS or from an authenticator app, in the same second it is entered. Securing an exchange account therefore means two things: choosing a second factor that survives both attacks, and hardening the routes around it, email, carrier, and withdrawals, in an order that cannot lock you out.
How a SIM swap actually works
A SIM swap never touches your phone. The attacker convinces your mobile carrier to move your phone number onto a SIM card they control. From that moment, your calls and texts, including one-time login codes, go to them instead of you. It is an attack on the carrier's customer process, not on your device, which is why a new phone or a careful password does nothing against it. The usual sequence: personal details are gathered from data leaks or social media; the attacker calls or visits the carrier pretending to be you, claims the phone was lost, and asks for the number on a new SIM; in some cases a carrier employee is bribed to make the switch. Moving the number to a new provider, a port-out, works the same way, and eSIMs change nothing, because the attack moves the number, not the plastic.
The reach of the swap is exactly the set of accounts that accept a text message as proof of identity: an exchange account with SMS two-factor or SMS password reset, and, more dangerous, an email account recoverable by SMS, because email resets almost everything else. What a swap cannot reach directly is a self-custody wallet whose seed phrase is stored offline on paper: no phone number can reset it. The trap is storage, a seed phrase photographed into a cloud gallery or saved in email is protected only by that account's login, and that login may be exactly what the swap opens. Reported SIM-swap losses have been falling. FBI complaint figures run from 2,026 reports and $72.7 million in losses in 2022 down to 971 reports and $17.4 million in 2025, but declining is not disappeared, and phishing, the attack that defeats typed codes, was the most frequently reported complaint type in 2025, with 191,561 reports.
Four ways to prove it is you, against three attacks
US digital-identity guidance sets the baseline: NIST SP 800-63B-4 (July 2025) classifies one-time codes delivered over the telephone network as a restricted authenticator, the only method in that category. A phone number is an administrative record held by a carrier, not a cryptographic secret, and the code leaves your control the moment it enters the mobile network. Services that still use it are expected to document a risk assessment, offer at least one non-restricted alternative, and watch for indicators such as a recent SIM change or number port. The four tiers, graded against the three attacks that actually take exchange accounts:
| Method | Stops password theft | Stops a SIM swap | Stops real-time phishing | Where the secret lives |
|---|---|---|---|---|
| SMS / voice code | Yes | No | No | Your phone number |
| Email code | Yes | Only as strong as the email account | No | Your email account |
| Authenticator app (TOTP) | Yes | Yes | No | An app on your device |
| Passkey / FIDO2 security key | Yes | Yes | Yes | A physical key or your device's secure hardware |
The table's last column explains the middle two. An authenticator app generates its code on the device itself, from a shared secret and the current time. The code is never delivered over a third-party channel and cannot be intercepted in transit, so a SIM swap runs into a dead end against it, a genuine, large improvement, and the right minimum for any exchange account. Its blind spot is the code itself: whatever can be typed can be relayed. A fake page that looks like the exchange passes the password and the six-digit code to the real exchange in the same second, and is inside. The code was valid; it was just not entered in the place its owner assumed.
Why only passkeys refuse a fake site
Passkeys and FIDO2 security keys solve the phishing problem not with a better code but by removing the code altogether. At setup, the device generates a key pair: the public half goes to the exchange, the private half stays on the device or the hardware key. At login, the device signs a challenge from the exchange, and the signature is firmly tied to the domain the key was created for. The operating system and browser enforce that binding, so a cloned page at a look-alike address is asking into the void: the device simply never answers it. There is nothing to type, hence nothing to relay.
Two designs need telling apart. A device-bound passkey or hardware security key never lets the private key leave the device, the strongest protection, at the price that a second key as a spare is not optional. A synchronised passkey lives in the keychain of a platform or password-manager account and follows the user across devices; its phishing protection is identical, because it rests on the same domain binding, but its overall security then rests on how well that sync account itself is protected. Either design defeats the SIM swap for the same reason TOTP does, nothing travels over the phone network, and both defeat the relay attack that TOTP cannot.
The setup procedure, in an order that cannot lock you out
The most common damage from two-factor setup is self-inflicted: a single method registered on a single phone, then the phone is lost. The order below prevents it. Treat each step as a gate, do not advance until it passes.
- Secure the email account first. Email is the master key: reset links for the exchange and nearly everything else land there. Give it a unique password and a non-SMS second factor (authenticator app or passkey), and remove SMS as its recovery method where the provider allows it. Gate check: you can sign in to email with no text message involved at any step.
- Spare first, switch second. Before changing anything on the exchange, set up the spare: a second hardware key, or a second device carrying the same authenticator app. Gate check: the spare produces a working code or completes a test sign-in.
- Print the recovery codes. Most exchanges display recovery codes exactly once. Put them on paper, store the paper somewhere other than with the device, and never store them as a screenshot in a photo library, a library the email account may expose. Gate check: the paper exists and is legible.
- Register the strong method on the exchange. Add a passkey or security key where offered; otherwise an authenticator app. Where the exchange allows several methods side by side, register both the passkey and TOTP, the pair beats either alone. Sign out and sign back in to prove it works before continuing.
- Only then remove SMS. As long as SMS stays active it remains a route into the account, and attackers always take the weakest active route. If the exchange will not let SMS be removed, remove the phone number as a recovery option wherever that is separately possible, and note the residual risk plainly.
- Allowlist withdrawal addresses. Add your own withdrawal addresses to the exchange's allowlist if it offers one. Even a successful break-in then leads nowhere while the destination address has not been approved, the layer that converts an account breach from a theft into an annoyance.
- Lock the number at the carrier. Ask the carrier to turn on its account lock against SIM changes and port-outs and set a separate carrier account PIN that is not a birthday or address fragment. In the US, rules in effect since 2024 require carriers to offer the lock free, authenticate securely before moving a number, and notify you before a SIM change goes through; outside the US, ask what equivalent PIN or port-out block exists.
What success looks like: sign-in requires a factor no text message can deliver; a working spare exists and has been test-used; recovery codes exist on paper; SMS is gone or explicitly noted as a residual route; withdrawals can only go to pre-approved addresses; and the carrier cannot move the number on a phone call alone. Six checks, each verifiable in the account's own settings screens.
If the phone goes dark: the first hour
The clearest sign of a swap is the phone suddenly showing "No service" or "SOS only" in a place where it normally works, often alongside a SIM-change notice or password-reset emails that were never requested. Speed matters more than completeness, because confirmed blockchain transactions cannot be reversed.
- Call the carrier from another phone. Ask for the number to be restored and the account locked. This call is first because every other recovery may depend on the number.
- Secure email from a trusted device. Change the password and sign out all other sessions, email is the reset path an attacker uses next.
- Lock the exchange accounts. Use any emergency freeze or account-lock the exchange offers, then change passwords and re-issue the second factor.
- Check any wallet whose backup touched an account. If a seed phrase may have been exposed through email or cloud storage, move remaining funds to a new wallet created on a clean device and stop using the old one.
- Report it. Carrier fraud team and local police, with times, notices, and transaction IDs recorded while they are fresh.
FAQ
Is SMS two-factor better than nothing? Yes, clearly: a stolen password alone is no longer enough. SMS is the weakest tier rather than a useless one, but for an account holding meaningful value it should be a fallback, not the plan.
Does an exchange account need 2FA if holdings sit in a hardware wallet? The wallet's coins have no account and no login, so 2FA does not apply to them, the recovery phrase is their protection. The exchange account still holds balances in transit, order history, and identity documents, and still deserves the full procedure above.
What if the exchange offers no passkey option? Use an authenticator app as the primary factor, complete every other layer, spare, printed codes, allowlist, carrier lock, and treat any login link arriving by message as hostile: navigate to the exchange by typing its address, since the remaining exposure is precisely the real-time phishing relay that only domain-bound factors defeat.
Are synchronised passkeys safe enough? Their phishing protection equals a hardware key's, because it rests on domain binding. The risk shifts to the sync account, which then needs the same treatment given to email above.
Reminder: Crypto Investing is educational content only, not financial advice. Crypto assets are volatile and can lose all value. Product terms, fees, and rules change, read the current prospectus, terms, or manufacturer documentation for any product you are considering.