Stay up to date with AI tools, model releases, and developer workflows that matter.

In a recent development, weekly. Free. One click to leave.

7 Day Free Trial. Cancel Anytime.

In a recent development, most identity theft online starts with a password. Someone reuses one, a site gets breached, a fake login page collects a few more, and suddenly a stranger is inside an email account resetting everything else.

According to recent reports, passkeys are the industry’s attempt to remove that weak link. Apple, Google, and Microsoft all support them now, and plenty of massive sites already offer them at sign-in. But do they actually make identity theft harder, or do they just move the problem somewhere else?

It has been revealed that the honest answer is a bit of both. This article looks at how passkeys work under the hood, which attacks they shut down, which ones they don’t touch, and what you need to get right if you’re adding them to your own site.

Industry observers note that passwords have one basic flaw. They’re a shared secret. You know it, the server knows it (or a hash of it), and anyone who gets a copy can use it from anywhere.

That one flaw feeds most of the account takeover attacks you hear about:.

Two-factor authentication helps, but not as much as people think. SMS codes can be intercepted or stolen through SIM swaps. One-time codes from an app can still be phished, because a fake site can simply ask for the code and pass it along in real time.

Industry observers note that A passkey is a pair of cryptographic keys made for one specific website. It’s built on the WebAuthn standard from the W3C and the FIDO Alliance.

Here’s the short version of what happens:.

Nothing secret ever crosses the network, and nothing secret is stored on the server. That small change has big consequences.

It has been revealed that this is the biggest win. Every passkey is tied to the domain it was created for. Your browser won’t use a passkey for example.com on examp1e-login.com, no matter how real the page looks. That protection only holds if the attacker can’t fall back to a password or a weak recovery path, which is why the rollout advice later in this article matters.

With a password, the user is the one who has to spot the fake. With a passkey, the browser checks the domain for them. People get tired and click bad links. The browser doesn’t.

In a recent development, each passkey is unique to one site, so there’s nothing to reuse. An attacker can’t take a passkey from one breach and try it somewhere else, because there’s no shared value to try.

Recent findings suggest that if an attacker dumps a site’s database, they get public keys. A public key can check a signature, but it can’t make one. So the stolen data can’t be used to sign in as anyone.

This doesn’t make a breach harmless (more on that below), but it does remove the part that usually leads straight to account takeovers.

Sources indicate that this is where a lot of articles get too optimistic. Passkeys protect the login step. Identity theft is a much bigger problem than that.

Every site needs a way back in for people who lose their phone. If that recovery path is an SMS code, an email link, or a support agent who can be talked into resetting things, attackers will go after that instead.

A site with passkeys and a weak recovery flow is only as strong as the recovery flow. This is the part developers most often underestimate. The OWASP Forgot Password Cheat Sheet is a good checklist for hardening it.

Passkeys do nothing about this kind of identity theft. It’s why credit monitoring, freezes, and alert services are still worth knowing about even when your logins are locked down. If you want a broader look at how those tools compare, these theft protection insights from Cybernews cover what different services actually monitor.

According to recent reports, someone can still call a person, pretend to be their bank, and talk them into approving something. Passkeys make that harder because there’s no code to read out loud, but scammers adapt. User education still matters.

Recent findings suggest that the good news is that the browser side is fairly small. Most of the real work happens on the server. The MDN guide to the Web Authentication API covers every option in detail, and the Passkeys Developer Resources site is a good source for implementation guidance.

Here’s a stripped-down registration call. In a real app, the challenge and user ID come from your server, never from the browser.

A couple of details worth noticing. The rp.id is what ties the passkey to your domain, so set it to the domain people actually sign in on. And user.id should be a random value rather than an email, since it’s stored on the user’s device.

Industry observers note that passkeys feel best when they show up right in the username field’s autofill menu. This is called conditional UI, and it lets you add passkeys without changing your login form much.

According to recent reports, users who have a passkey see it in the dropdown. Users who don’t can keep typing a password like before. That makes a gradual rollout much easier.

Sources indicate that verifying WebAuthn responses means checking the challenge, the origin, the relying party ID, and the signature. It’s very easy to get one of those subtly wrong. Use a well-maintained library instead of writing it yourself. SimpleWebAuthn is a popular choice for Node.js, and most major languages have solid options.

In a recent development, one thing that trips people up: signature counters. Older hardware keys increment a counter on every use, and some guides tell you to reject sign-ins if it goes backward. Synced passkeys often just report zero, so a strict counter check can lock out real owners.

A few practical tips from teams that have already done this:.

Recent findings suggest that there’s also the question of moving passkeys between platforms. Early on, passkeys stuck to one ecosystem, which annoyed a lot of people. Password managers like 1Password and Bitwarden now store and sync them across platforms, and the FIDO Alliance has published specifications for moving passkeys and other credentials between providers, the Credential Exchange Format and its companion protocol. It’s getting better, but it’s worth telling consumers where their passkeys live.

Yes, and by a lot, for one specific part of the problem.

In a recent development, passkeys close off phishing, credential stuffing, and password database leaks, which are the most common ways criminals take over accounts. That alone makes them one of the most meaningful security upgrades the web has seen in years.

But identity theft is bigger than logins. Weak recovery flows, stolen sessions, and leaked personal data all still lead to real harm, and passkeys don’t fix any of them on their own.

If you’re building for the web, treat passkeys as a strong foundation, not the whole house. Add them, make recovery just as solid, keep your sessions protected, and collect as little personal data as you can. That combination does far more for your users than any single feature.

It has been revealed that 7 Day Free Trial. Cancel Anytime.

→ Read original article