Fake IT Help Desk Call Scam: How Vishing Bypasses MFA and How to Stop It

Reviewed by IT Cares technicians · Updated November 16, 2026

Office desk with a landline phone handset off the cradle beside a laptop showing a login window and a warning triangle, illustrating a fake IT help desk call
A call from IT that asks you to sign in through a link it supplies is the warning sign.

Quick fix: 5 steps if you just got that call

  1. Stop. Do not click, sign in, approve a prompt or read out a code.
  2. Hang up and call IT yourself using a number from your directory, not the caller ID.
  3. If you already typed or approved anything, tell IT now so sessions and MFA devices can be revoked. See the first-hour response.
  4. Write down the time, number and web address, and report it (who to report to).
  5. Owners: adopt the verification script and block external Teams contact.

What we know and what is not confirmed yet (as of November 16, 2026)

Confirmed on the official page: On October 28, 2026, the Canadian Centre for Cyber Security (Cyber Centre) published Alert AL26-028, "Ongoing Help Desk Voice Phishing Campaign Targeting Canadian Organizations." It says the Cyber Centre is aware of an ongoing voice phishing (vishing) campaign against Canadian organizations, including organizations in the legal and financial sectors, where callers impersonate internal IT support and direct employees to fraudulent sign-in pages. It also says the campaign bypasses conventional multi-factor authentication (MFA) through real-time relay and frequently rotates its lure domains.

Reported in the news, not verified by us: Several Canadian law firms have been publicly reported as victims of social engineering incidents in the same period. We have not confirmed from a primary source which firms were hit through this specific technique, so we do not name them. Treat any claim that links a named firm to this exact campaign as unconfirmed unless that firm or the Cyber Centre says so.

Not confirmed: how many organizations were affected, how much data was taken, and whether your own organization is a target. The alert does not publish victim counts.

How the fake IT help desk call works, step by step

The scam is simple, and that is why it works. According to the Cyber Centre's alert, callers pretend to be internal IT or help desk staff and claim that something urgent must be done: a passkey migration, an MFA enrolment, or a password change. The call can reach a personal mobile phone, an office phone, or a Microsoft Teams account, and the caller ID may be spoofed to look like your real help desk. The caller may mention the names of real IT staff or a missed deadline to sound credible, and, as the alert notes, may hang up if the employee tries to verify the call independently.

Here is the sequence, written as a defender would see it.

  1. The pretext. "Hi, this is Marc from IT. We are moving everyone to passkeys this week and your account is flagged as not migrated. If it is not done by end of day, you will be locked out."
  2. The link. The caller tells you to open a web address, often read out slowly or pasted into a Teams chat. The page carries your company's name and looks like a normal enrolment or single sign-on portal. The alert gives the pattern: your organization's name may appear as a subdomain of a generic domain, or be built into a newly registered domain, combined with words such as "sso", "mfa", "passkey", "signin", "portal", "setup" or "enroll".
  3. The relay. You type your username and password. The fake page passes them to the real Microsoft or other identity provider in real time. The real provider then asks for your second factor, the fake page shows you the same prompt, and you approve it or type the code. The alert describes this as an adversary-in-the-middle page that relays credentials and MFA responses to the legitimate identity provider, giving the attacker an authenticated session.
  4. The coaching. The caller stays on the line and talks you through each prompt: "You will get a code, read it to me or enter it on the page." That live coaching is the difference between this scam and ordinary phishing email.
  5. The takeover. With an authenticated session, the alert says the actor has been reported to register attacker-controlled MFA methods or devices to keep access, delete password-reset, MFA-change and security notification emails, reset passwords for non-SSO applications, and use scripts and Microsoft Graph to pull data out of Microsoft 365 and connected apps, including SharePoint and OneDrive. Extortion demands follow after data theft.

Notice what is missing: no malware, no attachment, no exploit. Antivirus has nothing to catch. The attacker uses your own legitimate login, which is why this is a people-and-process problem first and a software problem second.

Why this beats ordinary MFA (and what does not)

Most small businesses turned on MFA because they were told it stops account takeover. It does stop a huge share of attacks: a stolen password alone is useless. But the common MFA methods (SMS codes, authenticator app codes, "approve" push notifications) all share one weakness: a human can be talked into handing over or approving them in real time. The relay page does not need to guess anything. It just forwards what you type.

The Cyber Centre's first recommendation is phishing-resistant MFA: FIDO2 security keys, device-bound passkeys, or platform authenticators. These tie the sign-in to the real website address. A passkey created for the genuine Microsoft sign-in page will not work on a lookalike domain, so there is nothing for the caller to relay. We explain the options in plain language in our business guide to passkeys and our general MFA guide for business owners.

MFA methodStops stolen password?Stops a live relay page?Can a caller talk you into it?
SMS codeYesNoYes: "read me the code"
Authenticator app 6-digit codeYesNoYes: "type the code on the page"
Push "Approve" promptYesNoYes: "tap Approve to confirm"
Push with number matchingYesMostly no, if the victim reads the number from the fake pageYes, with coaching
FIDO2 key or device-bound passkeyYesYes, bound to the real sign-in domainNo code exists to hand over

The table is our plain-language summary of how these methods behave in a relay attack, consistent with the Cyber Centre's advice to deploy phishing-resistant MFA. It is not a guarantee: a determined attacker can still try other routes, such as tricking someone into registering a new device. That is why verification of resets (covered below) matters as much as the technology.

One related trap: attackers also use the word "passkey" itself as bait. If a caller or message says you must "migrate to passkeys" through a link they supply, that is the lure, not the cure. Real passkey setup starts from your own account security settings, reached by typing the address yourself or using a bookmark. See our guide on fake passkey prompts in Microsoft 365 and the related change coming to SMS-based MFA in Microsoft 365.

Red flags: how to spot a fake IT call in the first 60 seconds

You do not need to be technical to catch this. You only need to know the pattern. Any one of the signs below is enough to stop and verify. Two or more means you should hang up.

Could the caller sound like someone you know?

The Cyber Centre's alert does not say callers use cloned voices, so we are not claiming that this campaign does. But voice cloning tools are widely available, and a call that sounds exactly like your IT manager is no longer proof of identity. This is another reason to rely on a verification procedure instead of your ears. Our article on AI voice cloning scams against businesses covers the wider problem.

Three realistic scenarios (illustrative, not real cases)

Realistic scenario: these are invented examples to show how the pattern plays out. They are not real incidents or real clients.

Scenario 1: The bookkeeper at a 12-person accounting office

At 4:40 on a Thursday afternoon, a bookkeeper receives a Teams call from an external account named "IT Helpdesk". The caller knows the office uses Microsoft 365 and mentions the name of the part-time IT contractor. He says everyone must "re-enrol" for passkeys before 5 p.m. or the mailbox will be suspended over the weekend. He sends a link to a page branded with the firm's name. She signs in, approves the prompt on her phone, and the page says "Done, thank you." Within the hour, the firm's shared client folders are being downloaded by scripts, and two forwarding rules appear on her mailbox. The firm only notices on Monday because a client calls about a strange email.

What would have stopped it: a rule that IT never asks for sign-in through a supplied link, a callback to the contractor's known number (the contractor knew nothing about it), and external Teams chat blocked by default.

Scenario 2: The office manager at a 30-person law practice

An office manager gets a call on her personal mobile from a number that matches the firm's help desk. The caller says a security audit found her account "not protected" and walks her through an MFA enrolment page. She is suspicious, and says she will call back. The caller replies: "That will take longer and your access will be suspended in a few minutes." She hangs up anyway, calls the real help desk from the company directory, and learns there is no such project. She reports the call using the practice's reporting channel, and the practice blocks the domain and warns everyone within the hour.

What worked: she followed the Cyber Centre's advice to verify independently, regardless of caller ID or urgency, and the practice had a simple place to report it. The alert recommends giving employees a clear channel for reporting suspicious calls and associated websites.

Scenario 3: The sole owner of a small contractor firm

A contractor with no IT department gets a text, then a call: "This is Microsoft 365 support, your subscription has a security problem, we need to enrol your account in a new passkey." The owner signs in on the fake page and reads out the code. By evening, his invoicing mailbox has a hidden rule forwarding messages containing the word "invoice" to an outside address, and a client has received a payment-change email that looks like his. The business loses a week and a $9,000 payment before it is caught.

What would have helped: understanding that Microsoft does not phone small businesses to demand sign-ins, a second person approving banking changes by phone, and a quick call to a trusted IT provider. Payment-change fraud like this is covered in our guide to business email compromise.

Office worker on a phone call looking at a laptop sign-in page with a concerned posture

The 30-second verification script for staff

Print this and tape it next to the phone, or paste it into your staff handbook. It works for any unexpected call, chat or text claiming to be IT, a vendor, a bank or a manager.

If someone calls about your account, do this

  1. Do not click, type, read out, or approve anything while they are on the line.
  2. Say: "I will call you back through the company directory." Then hang up. If they object, that is your answer.
  3. Call IT yourself using the number from the internal directory, your contract, or a number saved in your phone before the call. Never use a number from the message or the caller ID.
  4. Ask IT: "Did you just contact me about passkeys, MFA or my password?" A real project will be confirmed by a second channel.
  5. Report it to the person named in your policy, even if you did nothing, so others can be warned. Write down the time, number, and any web address.
  6. If you already clicked or typed something, tell IT immediately. Speed matters more than embarrassment. Nobody is in trouble for reporting.

Two extra tips make the script stick. First, tell staff that a real IT department will never be offended by a callback. Second, agree on a code word or a question that only the real IT person can answer (for example, the name of the ticket system). Do not reuse words that are posted on your website or social media.

Write an MFA reset and account recovery policy (one page is enough)

The Cyber Centre recommends strengthening verification for MFA resets, account recovery and new device enrolments. In a small business this is the single most valuable paper control, because the attacker's real goal is to get a new MFA device added to the account. If your help desk (internal, outsourced, or the owner) will reset MFA for anyone who calls with a good story, the whole security design collapses.

A workable one-page policy has five rules:

  1. IT never asks for a password, a code, or an approval over the phone, chat or text. Put it in writing and tell everyone. This rule alone defeats most of this scam.
  2. IT never sends sign-in links for enrolment. Passkey and MFA changes are done from the user's own security settings page, reached by a bookmark the company provides.
  3. MFA resets need two proofs. For example: a callback to the number on file in HR records and approval from the employee's manager by a separate channel. For executives and finance staff, require in-person or video verification with a known face.
  4. Announce planned changes in advance through a known channel, with the date and the exact steps. If a real migration is happening, users should hear about it before the first call.
  5. Report fast, no blame. A single reporting address or phone line, and a promise that reporting a mistake is always better than hiding it.

If you outsource IT, send the policy to your provider and ask them to confirm in writing that they follow it. Ask also how they verify a caller who claims to be you. If the answer is "we recognize the voice," that is not a control.

Technical controls, in order of impact

The Cyber Centre's alert lists several suggested actions. Here they are translated into a practical order for a company of 5 to 50 people on Microsoft 365. Menu names inside Microsoft admin portals change from time to time, so treat the paths below as a guide and confirm in your own tenant or ask your IT provider.

1. Move to phishing-resistant MFA, starting with the people who matter most

Begin with owners, partners, finance staff, and anyone with admin rights. Options are FIDO2 security keys, device-bound passkeys, or platform authenticators such as Windows Hello or Touch ID on managed devices. Keep two registered methods per person, and a documented break-glass admin account stored offline, so a lost key does not lock out the company. Our passkey guide for business explains rollout.

2. Block or limit external Teams contact

The alert recommends restricting external communications in collaboration platforms, including direct messages, calls and meeting invitations, to trusted or operationally required external users. In Microsoft Teams, Microsoft's documentation explains that external access has both an organization-level setting and user policies, and that both must allow it for a user to chat or meet with outside people. In the Teams admin center this is under the external access settings (look for Users, then External access). Many small businesses can switch to "allow only specific external domains" or turn external access off, then add the partners they actually use. Review guest access at the same time.

3. Require managed devices and restrict risky sign-in sources

The alert recommends conditional access that requires managed devices and restricts sign-ins from commercial VPNs, hosting providers and residential proxies. Conditional access is a Microsoft Entra feature that depends on your licence level, so check what your plan includes before promising yourself a configuration. A simple, high-value rule is: "Require a compliant or registered device for access to Microsoft 365 apps." Even if the attacker has your password and your relayed MFA, their own laptop does not meet the rule.

4. Shorten sessions and watch for new MFA devices

The alert suggests reducing session lifetimes and monitoring for suspicious MFA registrations, scripted or high-volume Microsoft 365 file access, security-email deletion, and newly registered domains that impersonate your organization. For a small business, the practical version is:

5. Make reporting easy

Set up a dedicated mailbox, for example a "report phishing" address, and a phone number staff can call. Post it in Teams and on the lunchroom wall. The best control in the world fails if the employee who received the call has nobody to tell.

Comparison: what each defence costs you and what it stops

DefenceEffort for a small businessWhat it stopsWhat it does not stop
Staff verification script and callback ruleLow: one hour of trainingMost of the live calls, because the attack depends on persuasionA very convincing attacker with real internal details
Written MFA reset policyLow: one pageAttackers trying to add their own device by calling the help deskMistakes when staff skip the policy
Block external Teams contactLow to medium: a settings change plus an allow-listCold chats or calls from outside accountsCalls to mobiles or office lines
Passkeys or FIDO2 keysMedium: hardware, enrolment, supportCredential and MFA relay on lookalike sitesFraudulent device registration if resets are weak
Managed device requirement (conditional access)Medium: licences and configurationSign-ins from the attacker's own machineAn attacker using a compromised employee device
Monitoring and alertsMedium: ongoingLate detection, shortens dwell timeThe first successful login

The cheapest items on this list also have the biggest effect. If you can only do three things this week, do the verification script, the reset policy and the external Teams restriction. They cost almost nothing and each addresses a step in the attack chain described by the Cyber Centre.

Illustration of a telephone connected by a dotted line through a fishing hook to a laptop with a padlock

If someone already gave access: the first-hour response

If an employee entered a password, approved a prompt, or read out a code to a caller, assume the account is compromised until proven otherwise. Do not wait for evidence of damage. The attacker described by the Cyber Centre moves quickly: it registers its own MFA device, hides the security emails, and starts copying files with scripts. The first hour decides how bad this gets.

The Cyber Centre's advice, if compromise is suspected, is to revoke active sessions and tokens, reset credentials, remove unauthorized MFA methods or devices, and review mailbox and file activity. Here is that advice as a numbered response, in order.

  1. Stay calm and keep the employee reachable. They are the best witness. Ask exactly what they typed, which address they visited, when, and what the caller said. Write it down with times.
  2. Cut the attacker's session. An administrator should sign in from a clean device and, for the affected user, revoke all sessions and sign-out tokens. In the Microsoft Entra admin center, look on the user's page for the option to revoke sessions. Microsoft's own documentation on revoking user access in Microsoft Entra ID explains that access tokens last about an hour by default and that revoking a session is part of the process, so do this even if you also reset the password.
  3. Reset the password from a clean device. Use a new, long, unique password or passphrase. If the account is also used for other apps that do not sign in through single sign-on, reset those too, because the alert says the actor resets passwords on non-SSO applications.
  4. Review and remove authentication methods. Open the user's authentication methods and delete any phone, authenticator app or security key you do not recognize. Then make the real owner re-register methods, ideally a passkey or hardware key, in a verified setting such as in person or on a video call.
  5. Look for inbox rules, forwarding and delegates. Check mailbox rules, external forwarding, delegated access, and recently created app consents. Attackers use forwarding rules to keep watching the mailbox and to hide security notices.
  6. Check what was accessed. Review the Microsoft 365 audit log for the account: file downloads, SharePoint and OneDrive access, mail items deleted, and any sign-ins from unusual places. Preserve the logs before anything is changed or ages out. The alert specifically recommends preserving identity provider and Microsoft 365 audit logs for investigation.
  7. Warn your contacts if mail was sent from the account. If fraudulent messages may have gone to clients or suppliers, tell them by phone, not by email from the same account. Be ready for follow-up payment-change requests.
  8. Block the lure domain. Add the web address the employee visited to your email and web filters, and report it (see the reporting section below).
  9. Check money movement. If anyone in finance was involved, call your bank to flag recent changes to payee details. Combine this with our guide to what to do when an employee clicks a malicious link.
  10. Decide whether this is a bigger incident. If data was copied or you receive an extortion message, shift to the larger response: see what to do in the first 24 hours after a ransomware attack (the early steps are similar even if no ransomware was deployed). Do not pay or negotiate before you have taken advice.

Checklist: copy this into your incident notes

  • ☐ Employee interviewed, timeline written down (time of call, number, link, what was entered)
  • ☐ All sessions and tokens revoked for the user
  • ☐ Password reset from a clean device (and non-SSO app passwords reset)
  • ☐ Unknown MFA methods and devices removed, real methods re-registered
  • ☐ Mailbox rules, forwarding, delegates and app consents reviewed
  • ☐ Audit logs exported and saved
  • ☐ Lure domain blocked and reported
  • ☐ Clients, suppliers and bank warned if needed
  • ☐ Privacy-breach assessment started (see below)
  • ☐ Lessons learned: policy and training updated within 7 days

Who to report to in Canada

Reporting helps you and everyone else. It gives investigators the infrastructure details that let them block the next wave.

Train staff in 20 minutes, not 2 hours

Annual awareness videos rarely change behaviour. What works is a short, concrete drill, repeated. Here is a 20-minute format for a team meeting.

  1. Five minutes: the story. Tell the Cyber Centre alert in plain words: callers pretend to be IT, send you to a page with your company's name, and relay your login and code. Mention that legal and financial organizations have been targeted, so every professional services firm should assume it applies to them.
  2. Five minutes: the three rules. IT never asks for a code. IT never sends sign-in links. We always call back through the directory.
  3. Five minutes: a live drill. Have a manager play the caller, using a harmless script, and give each person a turn to answer with the callback line. Practising the sentence "I will call you back through the directory" makes it easier to say under pressure.
  4. Five minutes: the reporting path. Show exactly where to report, and promise no blame.

Repeat the drill every quarter with a different pretext (a payroll problem, a software licence, a "security audit", a shared document). The pretext changes, but the rule that you never act on an unverified call stays the same. For ideas on the wider programme, see the cybersecurity checklist for a 10-person company and the related guide to spear phishing.

Budget: what prevention costs versus what an incident costs (CAD)

Costs vary a lot with your licences, device count, and who does the work, so treat the figures below as rough ranges for planning, not quotes. They are our estimates, not vendor or government statistics.

ItemDIY or low costWith a technician
Written verification and MFA reset policyFree, 1 to 2 hours of owner time1 to 2 hours of consulting
Staff drill and reminder cycleFree, 20 minutes per quarterOptional facilitation
External Teams restriction and allow-listFree, uses existing admin accessAbout 1 hour of setup
FIDO2 security keysOften roughly $30 to $90 per key; buy two per personSetup and enrolment support
Conditional access and device managementDepends on your Microsoft licence tierHalf a day to several days depending on size
Incident response after a compromiseYour time, plus risk of missing somethingHours to days, plus legal and notification costs if data left

A single takeover that leads to a fraudulent payment, lost billable days, and client notifications can easily cost far more than a year of the preventive measures above. IT Cares offers a single 119.99$ CAD, 60-minute Expert Consultation if you want a technician to review your Microsoft 365 setup against the Cyber Centre's recommendations. You can also simply call us.

Physical security key plugged into a laptop USB port on a tidy desk

A 30-day plan for the owner of a small firm

If you also keep client files in a cloud account, remember that OneDrive and SharePoint are not backups, so a compromised account that deletes or encrypts files needs real restore points. See our Microsoft 365 backup guide.

Official resources

When to call a pro

You can handle the policy, the training and the Teams settings yourself. Call a professional if any of these apply:

IT Cares provides remote support across Canada and on-site service in Quebec, and has done so since 2014. We can review your Microsoft 365 sign-in and audit logs with you, remove unauthorized access, and set up the controls in this guide. Call 1 (888) 711-9428, or see our remote support page and our cybersecurity services for Canadian businesses. IT Cares is not affiliated with Microsoft or the Government of Canada.

The campaign may change. Domains rotate, and the caller's story can shift from passkeys to payroll to licence renewals. The defence does not change: independent verification, no sign-ins through supplied links, and phishing-resistant MFA.

Still stuck? Get a technician on it now

Remote support from IT Cares: we connect to your device, fix it with you watching, and explain what happened.

Frequently asked questions

What is a fake IT help desk call scam?
It is a voice phishing (vishing) attack where the caller pretends to be your internal IT or help desk and says you must complete an urgent task such as a passkey migration, MFA enrolment or password change. The caller sends you to a fraudulent sign-in page that steals your login. The Canadian Centre for Cyber Security describes this tactic in Alert AL26-028.
How does this scam bypass multi-factor authentication?
The fake page relays what you type to the real identity provider in real time. When the real provider asks for your MFA code or approval, the fake page shows you the same prompt and passes your answer along. According to the Cyber Centre, this gives the attacker an authenticated session even though MFA was enabled.
Which MFA methods resist this attack?
The Cyber Centre recommends phishing-resistant MFA: FIDO2 security keys, device-bound passkeys, or platform authenticators. They are bound to the real sign-in domain, so they do not work on a lookalike page. SMS codes, authenticator codes and simple approve prompts can be relayed.
Who is being targeted?
The Cyber Centre says the campaign targets Canadian organizations, including those in the legal and financial sectors, with October activity concentrated on the legal sector. Any organization using Microsoft 365 or similar single sign-on could receive these calls, so small firms in other sectors should still prepare.
Which Canadian firms were hit?
News reports mention several Canadian law firms dealing with social engineering incidents, but we have not confirmed from a primary source which firms were affected by this specific campaign, so we do not name them. The Cyber Centre alert does not list victims.
Can caller ID prove the call is from my IT team?
No. The alert states that callers may spoof legitimate help desk numbers. Always hang up and call back using a number you already trust, such as the internal directory or your IT contract.
What should I say if the caller pressures me not to verify?
Say that company policy requires a callback and end the call. The alert notes callers may end the call if the employee tries to verify independently, so resistance to a callback is itself a warning sign.
I already entered my password and approved a prompt. What now?
Tell IT immediately. An administrator should revoke your active sessions and tokens, reset your password from a clean device, remove any unknown MFA methods or devices, review mailbox rules and file activity, and preserve audit logs, as the Cyber Centre advises.
Do I need to report this to anyone?
The Cyber Centre asks recipients who discover matching activity to report via My Cyber Portal or contact@cyber.gc.ca. You can also report to the Canadian Anti-Fraud Centre and the police, notify your cyber insurer if you have one, and check whether privacy-breach notification rules apply to you.
Does a Quebec business have to notify anyone about this?
If personal information was accessed and there is a risk of serious injury, Quebec's private-sector privacy law has incident duties, including keeping a register and notifying the Commission d'accès à l'information and affected people. Confirm the current rules with the CAI or a lawyer. This is general information, not legal advice.
Should I block external Teams chats?
For many small businesses, yes. The Cyber Centre recommends restricting external communications in collaboration platforms to trusted or operationally required external users. In Teams, both the organization setting and user policies control external access, so you can limit it to specific domains.
Is a passkey migration request ever legitimate?
A real migration is announced in advance through a known channel and starts from your own account security settings, not from a link read over the phone. If a caller creates urgency and supplies the link, treat it as the scam.
Can IT Cares help us set this up?
Yes. IT Cares provides remote support across Canada and on-site service in Quebec. We can review your Microsoft 365 setup, restrict external Teams access, plan passkey rollout, and help investigate a suspected compromise. IT Cares is not affiliated with Microsoft or the Government of Canada.

Sources and official references

Last verified: November 16, 2026

Need Help?