10 min read
FIDO2 and Passkeys Make MFA Phishing-Resistant in Microsoft 365
Most of the Microsoft 365 accounts compromised in the last 18 months had MFA enabled at the time of the attack.
4 min read
Nicole Walker
:
Updated on May 7, 2026
Microsoft 365 is the productivity backbone for most modern organizations.
Email, identity, file storage, collaboration, and third-party applications all run through a single cloud identity. That consolidation is efficient. It is also predictable from a security standpoint.
As adoption grows, attackers do not need to innovate. They simply get more opportunities.
Incident response data shows the same pattern across compromised tenants. Business email compromise (BEC), token theft, OAuth abuse, and MFA fatigue attacks are not new threats. They are predictable outcomes of identity-first environments built on default configurations and outdated authentication assumptions.
For real attack patterns and how these compromises play out in active tenants, listen to our Demystifying Microsoft episode on Microsoft 365 account compromise.
The problem is not a lack of security controls. It is that identity exposure grows faster than those controls are enforced.
Microsoft 365 centralizes critical business workflows behind a single identity plane. A compromised account can grant access to Outlook, Teams, SharePoint, OneDrive, and connected SaaS applications in a single step.
From an attacker's perspective, this creates three clear advantages:
A large, consistent user base
Microsoft 365 tenants share similar identity patterns, authentication flows, and default settings. Techniques that work against one tenant often scale across thousands more with little added effort.
Global, browser-based access
Cloud access is not limited by network location or device ownership. A valid session token allows entry from virtually anywhere. This reduces the need for malware, lateral movement, or on-premises access.
Long-lived configurations and identity debt
Many tenants were deployed years ago and never revisited. Identity settings that were acceptable in earlier threat models remain are still in place, even as phishing infrastructure, token theft, and OAuth abuse have evolved.
Business email compromise is one of the most financially damaging attack categories year after year. The mechanics are simple, the success rate is high, and the risk to the attacker is low.
Modern phishing does not rely on credential theft alone. AiTM infrastructure can capture valid session tokens during legitimate authentication flow.
Here is how it works:
Password resets alone do not invalidate these sessions. Token revocation is required.
OAuth app consent is one of the quietest persistence paths in Microsoft 365. Users are tricked into approving apps that request access to mail, files, or directory data.
Once consent is granted, the app operates independently of the user's password and MFA state. These apps frequently survive remediation and continue accessing data until permissions are explicitly revoked.
Push-based MFA without additional context creates predictable failure conditions. Attackers send repeated authentication prompts until a user approves one, often during off-hours or moments of distraction.
Microsoft introduced number matching to address this. Many tenants still rely on legacy MFA experiences that remain vulnerable.
Even with MFA in place, password reuse is still effective. Credentials from unrelated breaches are routinely tested against Microsoft 365 tenants. When combined with MFA fatigue or token capture, reused credentials shorten compromise timelines significantly.
Attackers rarely move to fraud immediately. They establish persistence first, then observe.
Step 1: Establish persistence
Step 2: Monitor the mailbox
Step 3: Act at the right moment
Most of these attacks are financially motivated. The goal is to move money quickly or access data that can be monetized. In many cases, the compromise is not discovered until financial or legal consequences surface.
For a deeper breakdown of persistence mechanisms and how to confirm containment, check out our What Happens After a Microsoft 365 Compromise article.
Reducing compromise risk does not require major architectural change or added operational burden. The highest return comes from a focused set of controls that disrupt the most common attack paths.
The goal is not perfect security. It is meaningful risk reduction with minimal friction.
1. Strengthen Identity Protections
Phishing-resistant MFA (FIDO2 security keys, passkeys) eliminates token replay and AiTM phishing, especially for high-risk accounts
Number matching replaces blind push approval with intentional validation, significantly reducing MFA fatigue.
Risk-based Conditional Access ties trust to behavior, using sign-in risk, device state, and location to close gaps that static policies miss.
2. Restrict OAuth Consent and Persistence Paths
Limit user consent to verified publishers and low-risk permission scopes
3. Reduce Mail-Based Execution Risk
Disable automatic external forwarding by default, with documented exceptions applied deliberately
Configure Defender for Office 365 policies to address misconfigurations commonly associated with BEC
Password resets are not enough. You also need to review and revoke forwarding rules, registered authentication methods, OAuth app permissions, and active sessions.
Outbound spam policies can disable automatic forwarding tenant-wide while allowing narrow, documented exceptions where needed.
Yes, if number matching and contextual prompts are not enforced. Simple push approval remains vulnerable to push bombing.
Restrict user consent, enforce verified publishers requirements, and regularly audit enterprises application permissions.
Microsoft 365 account compromise is not driven by obscure exploits or zero-day vulnerabilities. It is the result of identity sprawl, long-lived tokens, and trust assumptions that no longer hold at scale.
The attack patterns are consistent. The remediation paths are already known.
Tenants that experience repeated incidents treat identity controls as optional configuration rather than foundational infrastructure.
As organizations expand their use of Microsoft 365, reducing compromise risk is less about adding tools and more about enforcing discipline where identity already matters most.
Related resources on Microsoft 365 Account Compromise
Listen and watch the Demystifying Microsoft episode on account compromise
10 min read
Most of the Microsoft 365 accounts compromised in the last 18 months had MFA enabled at the time of the attack.
14 min read
The average BEC attack costs organizations over $125,000. Most of them succeed not because defenses were bypassed, but because the right controls...
9 min read
Microsoft has introduced Microsoft 365 E7 as a new top‑tier enterprise license designed for organizations moving beyond AI experimentation. E7,...
1 min read
Most Microsoft 365 compromises follow a familiar pattern. Access beings through phishing or token theft. Persistence is added quietly. Then attackers...
3 min read
Most Microsoft 365 security incidents do not start with a traditional breach. They start with a legitimate login. An attacker gains access to a...
1 min read
In today's threat landscape, a username and password alone are no longer sufficient to protect your organization.