If a stranger on Teams says they’re from IT support and wants to “fix” something, treat it as a scam until you’ve proven otherwise. That’s my short answer. The longer one matters, because this attack is being used against businesses right now and it’s built to rush you.
What’s happening
Researchers at Arctic Wolf Labs have described what they call a “recurring social engineering pattern” across hundreds of cases. Attackers use Microsoft Teams’ ability to start a chat with people outside your company. They pose as IT help and try to get you to give them control of your computer. The details were reported by The Hacker News on October 26, 2026.
The reporting says the attackers keep reusing the same tricks:
- Throwaway Microsoft 365 accounts. They set up new or acquired Microsoft 365 tenants (basically, a company account) just for this. Roughly 30–40% of the cases Arctic Wolf looked at used an address ending in onmicrosoft.com, a default domain Microsoft hands out automatically. The names are chosen to sound official, such as “helpdeskteam365” or “itadmin922”.
- Official-sounding display names. Examples include “Help Desk,” “IT Service Desk,” “HelpDesk Manager,” and “Enterprise IT & Security (Internal).” Some use a friendly first name too. Lately, some add your company’s name or domain to look more real.
- Paid domains. Arctic Wolf also saw a recent shift toward paid domains. Of those, about half used .top and about 40% used a regular .com.
The point of those names is to get past the warning Teams shows when a message comes from outside your organization. A name that says “(Internal)” is designed to make you stop reading the warning.
The inbox flood that comes first
In many of these cases, the Teams message is preceded by what’s called mail bombing. Your inbox suddenly fills with hundreds of newsletter or subscription confirmations in a short time. You can’t find real email, and you’re annoyed and a bit panicked.
Then a “helpful” person from IT shows up on Teams offering to fix it. The timing is the trick. The flood is not a coincidence, and the person offering help is the one who caused it.
What they want from you
The goal is hands-on access to your computer. The report says attackers get victims to allow remote access through legitimate remote support tools. The ones named include Quick Assist, AnyDesk, ScreenConnect, TeamViewer, Atera, Zoho Assist, NetSupport, and LogMeIn. These are real products that real IT people use, which is why they work so well for this. Your antivirus has no reason to object to software that’s meant to give someone remote control.
Once they’re in, the report describes them looking around the network, installing a backdoor, moving to other machines (including the main servers that manage a company’s accounts), adding more remote tools as a backup way in, copying out sensitive data, and then deploying ransomware. Sophos has tied some of this activity to Chaos ransomware.
In one case Arctic Wolf described, the attacker told the victim to run a PowerShell command. That command downloaded an installer from Microsoft’s Azure cloud storage, which then ran a script and installed a backdoor that collected information about the company’s computers and users and sent it to an outside server.
One more detail worth knowing: the attackers generally avoid pasting a link into Teams, since Microsoft’s link scanning could catch it. Instead, they go to the malicious download themselves during the remote session, while you watch. So “there was no suspicious link” doesn’t mean the conversation was safe.
Arctic Wolf’s summary is blunt: this works because it “exploits trust rather than a software vulnerability.” There’s no bug to patch. The weak spot is a person deciding to believe a chat message.
How to tell if it affects you
The report says these attacks specifically go after organizations where external access is enabled in Teams. If you use Teams at work, even as a one-person business, it’s worth finding out whether yours is.
Ask yourself a few questions about the message in front of you:
- Do you actually have an IT department or provider? If you’re self-employed and nobody has ever told you “IT will contact you on Teams,” an unprompted message from “IT” is already suspect.
- Did it arrive right after your inbox filled with junk?
- Does the sender’s name say “Internal,” but Teams marks them as external? Click on the person’s profile and look at the full address. A name like “IT Support (Internal)” attached to an address you don’t recognize is a problem.
- Are they asking you to install something, open a remote tool, or paste a command? Real IT doesn’t need you to do that over a surprise chat.
Any one of these is enough reason to stop.
What to do today
- Don’t reply, and don’t install anything. Don’t open a remote support tool, don’t read out a code, and don’t paste commands someone gives you.
- Verify through a separate route. Call or message your actual IT provider or colleague using contact details you already have, not anything from the chat.
- Block and report. Use Teams’ options to report the message, and tell whoever looks after your Microsoft 365 account. In Canada you can also report fraud attempts to the Canadian Anti-Fraud Centre.
- Tell your staff. One short message works: “Nobody from IT will ever ask you to install remote software through a Teams chat from someone you don’t know. If it happens, stop and tell me.” Make sure they know the inbox flood is a warning sign, not an emergency.
- Look at your external access settings. If you manage Teams, check whether outside people can start chats with your staff. If you don’t need that, turning it off or limiting it to specific organizations you trust cuts off the route this attack uses. That’s standard hardening, and it’s the most direct fix for what the report describes.
- Decide which remote tools are allowed. If your company uses one remote support product, make sure staff know which one. A tool they’ve never seen before is a red flag.
Signs something already went wrong
If you’ve already chatted with them, don’t panic, but act fast. Warning signs include:
- You let someone connect to your computer, or ran a command they gave you.
- A remote support program is installed or running that you didn’t put there.
- Someone moved your mouse or opened windows on their own.
- Your computer was doing things in the background after the “support session” ended.
If any of that applies, disconnect the computer from the internet (unplug the cable or turn off Wi-Fi), but don’t just keep using it. From a different, clean device, change your Microsoft 365 password and any others you used on that machine. Then tell your IT provider immediately. In the cases described, attackers moved from one computer to others and stole data before the ransomware appeared, so the first hours matter.
When to get help
Get your IT provider or a security professional involved if you gave anyone remote access, ran a command, or aren’t sure what happened. An analyst reviewing this would normally check which remote tools are installed, look at sign-in activity on the account, and look for anything else that has been added to the computer. That work is better done by someone who does it regularly. You also don’t need to be certain before asking. A false alarm costs you a short conversation. A missed intrusion can cost a lot more.
Sources
Featured image: Image by StartupStockPhotos from Pixabay.


