Passkeys are designed to make online accounts safer. They use public-key cryptography instead of asking you to type a reusable password, and properly implemented passkeys are resistant to traditional credential phishing.
But there is an unexpected social-engineering problem: attackers can use the security reputation of passkeys as a phishing lure.
In September 2026, Microsoft Security Research documented active cloud intrusions in which attackers impersonated IT helpdesk staff and told users that they needed to update or enroll a passkey, multifactor authentication (MFA), or single sign-on (SSO). The passkey story was often only the pretext. Victims could then be directed into adversary-in-the-middle phishing or device-code authentication flows that allowed attackers to obtain access to the victim's cloud identity. Microsoft Security Research documented the campaign on September 9, 2026.
This creates an important distinction:
The passkey technology is not necessarily what is being broken. The attacker is abusing the user's trust in the passkey setup process.
This guide explains how fake passkey setup scams work, what warning signs to look for, how device-code phishing fits into the attack, and what you should do if you already interacted with a suspicious authentication request.
What Is a Passkey?
A passkey is a passwordless authentication credential based on public-key cryptography. Instead of remembering a password and typing it into a login form, your device or credential provider uses a cryptographic key to prove that you control the account.
A typical passkey system uses a key pair. The service receives and stores the public key, while the private key remains protected by the authenticator or credential provider. When you authenticate, the legitimate service sends a challenge and the authenticator produces a cryptographic response.
This design provides an important phishing-resistance property: a passkey is associated with the legitimate authentication origin rather than being a reusable text secret that you can simply copy into another website.
That is why passkeys are a major improvement over traditional passwords for supported accounts.
However, phishing-resistant authentication does not make social engineering disappear.
Why Are Attackers Pretending to Set Up Passkeys?
Passkeys are becoming more common, which means users are increasingly likely to recognize terms such as “passkey,” “security key,” “passwordless,” and “MFA enrollment.” Attackers can exploit that familiarity.
Instead of telling a victim:
“Click this link to give us your password.”
an attacker can make the request sound like a legitimate security improvement:
“Your company is upgrading authentication. You need to register your new passkey before your account is locked.”
The second message sounds much more believable, especially when it comes from someone claiming to be IT support.
Microsoft observed campaigns in which attackers used phone calls, SMS messages, and even trusted internal communication channels to make these requests appear legitimate. The campaign could then move the victim toward a phishing page or a device-code authentication flow. Microsoft's investigation found that attackers could subsequently add their own authentication methods and access cloud resources.
How the Fake Passkey Scam Works
The exact sequence can vary, but a common attack pattern looks like this:
- Target research: The attacker identifies employees, organizations, job roles, or public contact information.
- Social engineering: The victim receives a phone call, message, email, or chat from someone pretending to be IT support or another trusted person.
- Security pretext: The attacker claims that the victim must configure a passkey, MFA, SSO, or account-security update.
- Fake authentication experience: The victim is directed to a website that resembles a legitimate identity provider.
- Authentication interception: Depending on the attack, credentials and session information may be captured through adversary-in-the-middle phishing, or the victim may unknowingly authorize an attacker-controlled device-code request.
- Persistence: The attacker may add an authentication method they control.
- Cloud reconnaissance: The compromised identity may then be used to inspect applications, files, mailboxes, or organizational resources.
The critical point is that the attacker does not necessarily need to defeat the cryptography of a properly implemented passkey.
Instead, the attacker attempts to control the conversation around authentication.
Device-Code Phishing: The Part Many Users Don't Understand
Device-code authentication is a legitimate authentication mechanism. It exists for situations where a device or application cannot conveniently display a normal interactive login interface.
The problem occurs when attackers abuse the legitimate flow.
In a device-code phishing attack, the attacker can initiate an authentication request and provide the victim with a code. The victim is then persuaded to enter that code into a legitimate authentication page.
The page may actually belong to the real identity provider. That is what makes the attack confusing.
The victim may believe they are completing a normal sign-in, while the code is actually associated with an authentication request controlled by the attacker.
Microsoft has documented device-code phishing as an active technique and recommends blocking device-code flow where it is not required.
This is an important lesson for users:
A legitimate-looking authentication website does not automatically mean that the authentication request you are approving is legitimate.
7 Warning Signs of a Fake Passkey Setup Request
1. You did not initiate the setup
This is one of the strongest warning signs.
If someone suddenly contacts you and says you must create a passkey immediately, stop and verify the request through an independent channel.
Do not assume that an authentication request is legitimate merely because it mentions a real security feature.
2. Someone creates artificial urgency
Attackers frequently use time pressure because it reduces careful decision-making.
Be suspicious of messages such as:
- “Your account will be disabled in 10 minutes.”
- “You must register your passkey right now.”
- “Your MFA is expiring today.”
- “Complete this verification before your next login.”
A real security process may have deadlines, but unexpected urgency combined with a request to authenticate should trigger verification rather than immediate action.
3. The person contacting you asks you to use a code
Be especially careful when someone on the phone or in chat gives you an authentication code and asks you to enter it into a website.
Ask yourself:
Who initiated this authentication request, and which application am I actually authorizing?
If you cannot answer those questions, stop.
4. The URL is unfamiliar
Attackers can create domains that contain the victim organization's name while still being controlled by an unrelated domain.
For example, a suspicious URL could look like:
company-name.secure-passkey-example.com
The presence of your company name somewhere in the URL does not prove that the company owns the domain.
Pay attention to the actual registrable domain and not just the words displayed at the beginning of the address.
5. The request comes from an unexpected channel
If your organization's normal security process uses an internal portal, but someone suddenly tells you to complete authentication through a personal phone call or an unfamiliar messaging account, verify it.
Microsoft observed campaigns where phone calls and SMS messages were used as part of the social-engineering stage.
6. You are asked to install unusual software
A passkey normally does not require you to install random remote-access software because someone claiming to be IT told you to do so.
Requests to install remote administration tools, browser extensions, certificates, unknown applications, or configuration profiles should be independently verified.
7. The process does not match your organization's normal workflow
Security teams should have a documented authentication process. If the instructions you receive suddenly differ from normal procedures, do not treat that difference as proof that the organization is under attack or that the new procedure is legitimate.
Contact your organization's IT or security team using a known phone number, internal ticketing system, or official directory.
Legitimate Passkey Setup vs Fake Passkey Setup
| Signal | More consistent with legitimate setup | Potential warning sign |
|---|---|---|
| Who started it? | You initiated setup or expected the rollout | Someone unexpectedly tells you to start |
| Communication channel | Known company portal or established process | Unexpected phone call, SMS, or chat |
| URL | Expected official domain | Lookalike or unfamiliar domain |
| Urgency | Normal documented rollout | Threats of immediate account closure |
| Authentication code | Expected as part of your own initiated flow | Someone else gives you a code and tells you to enter it |
| Software installation | Known company-managed software | Unknown remote-access tool or extension |
| Verification | Confirmed through an independent channel | Attacker discourages you from contacting IT |
What Attackers Do After They Get Access
Getting through the initial authentication stage is often only the beginning.
Microsoft's September 2026 investigation described attack chains where compromised identities were followed by authentication-method changes, Microsoft Graph reconnaissance, SharePoint and OneDrive access, and mailbox activity.
This means an account takeover can quickly become a data-access problem.
An attacker may look for:
- Files stored in cloud drives.
- Emails containing sensitive information.
- Internal applications.
- Group memberships.
- Administrative roles.
- Additional accounts to target.
- Authentication methods that can be abused for persistence.
The attacker may also register an authentication method under their control. This is particularly dangerous because the original phishing event can end while the attacker retains a way to authenticate later.
What To Do If You Already Followed a Suspicious Passkey Request
If you think you interacted with a fake passkey or authentication request, do not wait to see whether something happens.
1. Stop interacting with the attacker
End the phone call or chat. Do not continue the setup process just because the attacker says you are almost finished.
2. Use a trusted device
If possible, access your account from a device you trust rather than continuing on the potentially compromised device or browser session.
3. Change the password if one was exposed
If you entered a password into a suspicious page, change it immediately through the legitimate service. If that password was reused elsewhere, change those accounts as well.
4. Review authentication methods
Check the account's security settings for newly added phones, authenticator applications, security keys, passkeys, or other authentication methods that you do not recognize.
5. Revoke active sessions
Where the service supports it, revoke active sessions or sign out other devices. For organizational accounts, contact the security or IT team so they can perform the appropriate session and token revocation.
6. Review recent account activity
Look for unfamiliar sign-ins, application authorizations, mailbox rules, file downloads, or other activity that occurred after the suspicious interaction.
7. Report the incident
If the account belongs to an organization, report the incident immediately through the organization's security process. Do not worry about “wasting their time.” Early reporting can give defenders a chance to revoke sessions and remove attacker-controlled authentication methods before the compromise expands.
How Organizations Can Reduce Fake Passkey Scams
Individuals are not the only line of defense. Organizations adopting passkeys should also design the rollout so that employees can distinguish legitimate enrollment from social engineering.
- Publish a clear passkey enrollment procedure.
- Tell employees which domains and applications are legitimate.
- Provide a verified channel for reporting unexpected authentication requests.
- Train helpdesk staff to avoid authentication workflows that weaken verification.
- Require phishing-resistant authentication for sensitive applications.
- Monitor for unexpected authentication-method registration.
- Review unusual device-code authentication activity.
- Apply risk-based access controls where appropriate.
- Restrict device-code flow when there is no business requirement for it.
- Monitor cloud activity for unusual combinations of sign-in, authentication-method changes, API reconnaissance, and large-scale data access.
Microsoft specifically recommends investigating unusual sign-ins together with new authentication-method enrollment, Microsoft Graph reconnaissance, and abnormal SharePoint, OneDrive, or mailbox activity rather than treating each event independently.
Do Fake Passkey Scams Mean Passkeys Are Unsafe?
No. That conclusion would confuse the authentication technology with the surrounding social-engineering attack.
A properly implemented passkey still provides an important defense against traditional credential phishing because there is no reusable password for a fake website to collect.
The problem is that attackers do not always need to break the authentication mechanism itself.
They can target the user before authentication, manipulate the authorization flow, compromise the endpoint, steal sessions, abuse recovery processes, or persuade the user to authorize something they did not intend to authorize.
That distinction matters because abandoning passkeys because of passkey-themed scams would remove a useful security control without solving the underlying social-engineering problem.
The correct response is to understand what the authentication request is actually asking you to authorize.
A Simple Rule for Passkey Safety
Remember this rule:
If you did not start the authentication process yourself, stop and verify who started it before approving anything.
This rule is useful because it does not require you to memorize every new phishing technique.
If an unexpected person tells you that your passkey, MFA, SSO, or account needs immediate attention, verify the request independently.
Passkey Scam Prevention Checklist
- Do not trust an unexpected passkey enrollment request.
- Verify IT or security requests through a known communication channel.
- Inspect the actual domain rather than relying on a familiar company name in the URL.
- Do not enter authentication codes supplied by strangers.
- Do not approve authentication prompts you did not initiate.
- Do not install unknown software because someone claims it is required for authentication.
- Keep your browser and operating system updated.
- Use phishing-resistant authentication where available.
- Review newly added authentication methods regularly.
- Know how to revoke sessions for your most important accounts.
- Report suspicious authentication requests quickly.
Final Takeaway
Passkeys are designed to make authentication harder to phish, but attackers are adapting by attacking the process around authentication.
Recent Microsoft investigations show how passkey-themed social engineering can be used as a believable pretext for AiTM phishing and device-code attacks. The attacker may pretend to help a victim configure stronger security while actually trying to obtain access to the victim's identity.
The safest approach is not to fear passkeys. It is to understand them.
When you receive an unexpected security request, slow down. Verify the person contacting you. Check the real domain. Understand what application or authentication request you are approving. Never enter a code simply because someone on the phone tells you to.
And if you already interacted with a suspicious authentication flow, treat it as a potential security incident: secure the account from a trusted device, review authentication methods, revoke sessions where appropriate, and report the incident.
The strongest authentication technology still depends on a user being able to recognize when an attacker is trying to manipulate the authentication process.
Sources
- Microsoft Security Research — Passkey-themed social engineering leads to identity and cloud compromise
- Microsoft Security Research — Unmasking EvilTokens: Getting to the root of device code phishing
- Microsoft Security Blog — Passkeys become the default authentication method in Microsoft Entra ID
- Microsoft Security Research — Storm-2372 conducts device code phishing campaign