Skip to main content
All incident reports
Incident report

Attackers pose as IT support on Microsoft Teams to win remote access

Attackers are impersonating internal IT staff over Microsoft Teams to persuade employees to hand over control of their machines, Microsoft has reported. The campaign sits alongside other impersonation cases the same article describes, including a US$25.6 million deepfake fraud at Arup and the 2025 breach at Marks & Spencer.

· First published

The short version

  • The Teams campaign. Microsoft reported attackers contacting employees through external Teams messages and calls while posing as internal IT personnel. Using pretexts such as an urgent technical problem, they persuaded staff to grant remote control through screen sharing or Quick Assist.
  • What follows access. Once connected, the attackers ran PowerShell commands to install malicious software, hold persistent access and gather information about the network. Microsoft also saw attempts to move towards domain controllers and certificate authorities.
  • Earlier cases. According to Hong Kong police, a January 2024 video conference involving an Arup employee was a deepfake; the employee subsequently transferred about HK$200 million (US$25.6 million). Marks & Spencer later confirmed its April 2025 disruption began with deceived employees at a third-party contractor, not a technical flaw in its own systems.
  • The common thread. The article argues that impersonation attacks succeed by abusing routine trust — password resets, support calls, third-party access — rather than technical exploits, and that verification processes must carry the load rather than employee judgement.

Timeline

DateEvent
January 2024According to Hong Kong police, a video conference involving a finance employee at Arup's Hong Kong office was a deepfake; the employee subsequently made transfers of about HK$200 million.
April 2025Marks & Spencer suffers an attack that disrupts operations for weeks; it later confirms entry came through social engineering at a third-party contractor.

The Teams impersonation campaign

Microsoft has reported a campaign in which attackers posed as internal IT staff to reach employees. Contact came through external messages and calls in Microsoft Teams, and the attackers offered pretexts such as an urgent technical problem. They then asked employees to allow screen sharing or remote control through Quick Assist.

Once connected, the attackers used PowerShell to install malicious software, establish persistent access and collect information about the network. Microsoft also observed attempts to move laterally towards critical infrastructure, including domain controllers and certificate authorities.

Arup and Marks & Spencer

In January 2024 a finance employee at engineering firm Arup's Hong Kong office received email that appeared to come from the company's UK-based CFO requesting a series of transfers. The employee was initially suspicious but was convinced by a video conference featuring what appeared to be the CFO and colleagues. According to Hong Kong police, the meeting was a pre-recorded deepfake assembled from publicly available video and audio. The employee made multiple transfers totalling roughly HK$200 million.

In April 2025 British retailer Marks & Spencer was hit by an attack that disrupted operations for weeks: online orders were suspended, contactless payments and Click & Collect were affected, and customer data was stolen. The company estimated the hit to its 2025/26 operating profit at around £300 million before mitigation and later confirmed that attackers entered by deceiving employees at a third-party contractor rather than through a flaw in its own systems. The article links the attack to the Scattered Spider criminal group.

Why verification is the weak point

The article says these cases share a mechanism rather than a toolset: attackers exploit familiar trust signals, whether a recognisable face and voice or a routine request arriving through a sanctioned channel. Password resets and remote IT support are ordinary tasks that turn into entry points when the requester's identity is not checked.

Third-party providers widen the gap. An organisation may control its own systems tightly while having little visibility into how an external support team verifies callers and authorises sensitive actions. The article also argues that awareness training cannot be the only control, because attackers research their targets in advance so that requests look routine, and staff need reliable checks they can fall back on.

What is not known yet

  • Which group or individuals carried out the Teams impersonation campaign; Microsoft has not named one in the reporting summarised here.
  • How many organisations or employees the Teams campaign reached, and whether any of the attempted moves to domain controllers or certificate authorities succeeded.
  • Whether data was taken in the Teams campaign, beyond the information gathering about compromised networks that Microsoft observed.
  • When the Teams campaign began and when it was detected.
  • Whether the Arup, Marks & Spencer and Teams cases are connected in any way beyond technique; no source links them.

What to do

If your staff use Microsoft Teams

  1. Review whether users can exchange messages and calls with external Teams tenants, and restrict it where there is no business need.
  2. Tell staff that IT will not ask for remote control of a device through an unsolicited chat or call, and that such requests go to a known internal number for confirmation.
  3. Treat any request to install or run remote-support tools during a support conversation as a step to be verified, not obeyed.

If you run a service desk or approve payments

  1. Verify callers with trusted authentication methods rather than security questions or personal details an attacker could have gathered.
  2. Require independent confirmation through previously verified contact details before password resets, privileged access changes or transfers.
  3. Apply the same checks to self-service portals, internal chatbots and automated workflows that handle account recovery.

For everyone else

  1. Assume that a convincing impression of a colleague, and not a technical flaw, is the likely route in, and design verification around that.
  2. Roll out phishing-resistant multi-factor authentication on the accounts that can grant access to endpoints and infrastructure.
  3. Map which third parties can request access to your systems or authorise actions, and ask how they verify identity.

Update log

  • First published.

Sources

New incident reports by email

One email a week with the reports published since the last one.

Want us to look at your domain?

Give us one apex domain and we'll send back a written snapshot of what is already visible from outside. No call required.