call center

Can a Scammer Get Past MFA by Calling Your IT Help Desk?

You turned on multi-factor authentication (MFA). Your staff, or just you, approve a prompt on a phone every time you sign in to Microsoft 365. Good. But here’s the question I’d want answered next: what happens when someone calls your IT provider and says, “I’m locked out, can you reset me?”

If the honest answer is “they ask a couple of questions and then do it,” you have a gap. A scammer doesn’t need to beat your MFA. They only need to talk to the person who can turn it off or replace it.

MFA fatigue isn’t the only way around MFA

You may have come across the term “MFA fatigue attack.” That’s where an attacker keeps triggering sign-in prompts, hoping someone gets annoyed enough to tap Approve. It’s a real nuisance, and it’s worth knowing about.

The help-desk call is a different route to the same goal. Instead of wearing you down, the attacker goes around you. This is the one I want to focus on, because it’s easy to overlook and easy to fix.

What’s actually happening

A recent BleepingComputer article lays out the problem well. One note up front: it’s sponsored content, written by a vendor (Specops Software) that sells a product in this area. I’m using it for the description of the problem, not recommending any product.

The core point is simple. Companies have spent a lot on making logins harder to break into. But when an employee loses access, the service desk is often the one who resets the password or restores the account. And the way they check who’s calling is often still “knowledge-based” questions, like the name of your manager or your department.

That’s the gap. The strong front door has a weak side entrance.

Why those questions don’t prove anything

The article points out how much an attacker can collect before they ever pick up the phone: LinkedIn profiles, company websites, social media and leaked data. That can give them job titles, colleagues, office locations and personal details. With that, answering “who’s your manager?” is trivial.

The article also notes that NIST’s digital identity guidelines no longer recognize security questions as an acceptable authentication method. Its summary is the line I’d put on a sticky note: knowing information about someone is not proof of who they are.

It names a group called Scattered Spider as having made the service desk a preferred target. According to the article, they don’t try to bypass MFA directly. They impersonate employees and convince agents to reset passwords or register new MFA devices. That second part matters. If an attacker gets a new device registered, your MFA now works for them instead of you.

It’s getting harder to spot by ear

The article says generative AI is making these calls more believable and easier to scale: convincing stories, natural conversation, even imitated voices. It cites a 2025 Gartner survey in which 62% of organizations had experienced a deepfake attack involving social engineering or automated processes in the previous year. It also cites Verizon’s 2026 Data Breach Investigations Report, which found that social engineering by text and voice call had a 40% higher success rate than traditional email phishing.

I’d read those numbers as direction, not as a forecast for your business. But the conclusion holds up: a help-desk agent can’t be expected to decide by gut feel whether a voice is real. The article says verification should be a required step, not a judgment call based on how convincing someone sounds.

How to tell if this affects you

It affects you if you use any outside IT provider, or anyone, including a colleague or family member, who can reset passwords or MFA for your accounts. Ask them these questions:

  • When someone calls asking for a password reset or a new MFA device, how do you confirm who it is?
  • Is that process written down, or does it depend on who picks up the phone?
  • Can one person make that change alone, or does someone else have to approve it?
  • Does the real account owner get told when a reset happens?

If the answers are “we ask a few questions,” “it depends,” and “one person,” that’s your to-do list.

The rule to set with your IT provider

Here’s the rule I’d ask for, in plain terms:

No password reset or MFA change happens because of an inbound call alone. The provider verifies identity through a separate, pre-agreed check that the caller can’t talk their way around, and the agent isn’t allowed to skip it, no matter how urgent or friendly the caller sounds.

The article’s version of this is a fixed process: the agent finds the account, starts a verification step, and only does the reset after the person has passed it. Methods it mentions include email, SMS and third-party authenticator apps like Duo. The important part is that the check is consistent and required.

For a small business, here’s how I’d turn that into something practical. These are general good-practice suggestions, not steps from the article:

  1. Ban “security questions” as proof. Anything an outsider could look up online shouldn’t count.
  2. Call back, don’t trust the caller. The provider ends the call and rings the number they already have on file, not one the caller offers.
  3. Treat new MFA devices as the highest-risk request. A new device deserves more checking than a password reset, since it hands over the keys.
  4. Require a second person for sensitive changes. For a one-person business, that might mean the provider confirms with you through a channel you set up in advance.
  5. Send a notice every time. The account owner should get a message on their existing contact details when a reset or device change happens.
  6. Give agents permission to say no. The article stresses that agents should feel confident refusing or escalating. If your provider pressures staff to “just get the customer working,” that’s a problem.
  7. Write it down. A one-page procedure that both sides have agreed to beats an unspoken habit.

The inconvenience is real. Someone locked out on a Monday morning will be annoyed by a callback. The article does note that account lockouts cost time, and it puts a single help-desk password reset at around $70. That’s a reason to make the proper process smooth, not a reason to skip it.

What to do today

  • Email your IT provider the four questions above and ask for the answers in writing.
  • Agree on the verification rule, and ask them to confirm it applies to everyone, including you as the owner.
  • Tell your staff that they should never be offended when IT calls back, and that they should be suspicious of anyone who urges them to skip a check.
  • If you handle your own IT, apply the same rule to yourself. Don’t reset someone’s access because they sound right on a call or in a message.

Signs something already went wrong

When an analyst reviews account activity after a suspected takeover, these are the first things they look for. You can watch for them too:

  • A notice that your password was changed when you didn’t change it.
  • A new sign-in method or device appearing on your account that you don’t recognize.
  • Sign-in alerts from places or devices you don’t use.
  • You were suddenly logged out and your old MFA stopped working.
  • A colleague mentions that “IT” called them about you, or you hear of a call you didn’t make.

If any of these show up, tell your IT provider right away and ask them to check the account’s recent sign-ins and registered sign-in methods. Change the password from a device you trust, and ask whether anything was reset by phone around the same time.

When to get help

If you see signs above and your provider can’t tell you how the change was authorized, treat that as a serious warning. Ask them to review the account’s activity in detail. If your provider has no written process for verifying callers, that’s worth raising before anything happens, not after.

MFA is still worth having. It just needs the same standard applied to the people who can reset it.

Sources

Featured image: Image by Tumisu from Pixabay.

Leave a Comment

Your email address will not be published. Required fields are marked *