MFA protects logins, not identity ecosystems
Modern identity platforms include dozens of authentication and authorization paths. MFA typically covers only a subset. Attackers target what was left behind.
Most organizations believe MFA solved credential-based compromise. In practice, attackers rarely defeat MFA directly. Instead, they walk around it—through legacy authentication paths, session reuse, token abuse, and identity design decisions that were never revisited after MFA was “turned on.”
These breaches are quiet, persistent, and often discovered months later—if at all.
MFA dramatically improves security posture, but it is not a control boundary. Legacy protocols, OAuth flows, session lifetime decisions, and recovery processes routinely bypass MFA without triggering alarms. Organizations that treat MFA as a finish line often discover this only after damage is done.
Modern identity platforms include dozens of authentication and authorization paths. MFA typically covers only a subset. Attackers target what was left behind.
In practice, this often means “for interactive logins by humans,” not service accounts, legacy clients, token refresh flows, or delegated permissions.
When MFA isn’t challenged, defenders assume legitimacy. Attackers blend into normal cloud operations instead of triggering alerts.
These are not edge cases. They persist because disabling them risks breaking business processes—and because few organizations inventory identity flows end-to-end.
IMAP, POP, SMTP AUTH, and older SOAP-based APIs often bypass modern MFA enforcement. Even when “mostly disabled,” exceptions linger for scanners, devices, or migrations.
MFA is checked at session creation, not every action. Stolen session cookies, refresh tokens, or persistent OAuth grants allow long-term access without re-authentication.
Once a malicious or over-permissioned app is authorized, MFA is irrelevant. The app operates independently, often with fewer logging and alerting hooks.
Password resets, break-glass accounts, emergency access policies, and helpdesk overrides frequently have weaker MFA requirements—or none at all.
MFA bypass attacks don’t look like brute force or phishing failures. They look like successful authentication and legitimate API usage.
SOCs often key on failed challenges or push fatigue. Quiet breaches generate neither, so they never escalate.
Access often originates from cloud infrastructure or locations already associated with the organization or its vendors.
OAuth-based abuse shifts activity from user sign-ins to app calls, which many teams monitor less aggressively.
Teams stop looking for bypass paths because they believe MFA “solved” the identity problem. Attackers rely on that assumption.
MFA success metrics often create false confidence. The real risk lies in what MFA does not cover—and how long attackers can operate undetected.
Quiet breaches persist longer, increasing data exposure, regulatory risk, and recovery costs once discovered.
Legacy auth paths and undocumented exceptions are often discovered during incidents—or by regulators—not internal reviews.
Organizations invest heavily in MFA tooling while underfunding identity architecture reviews, lifecycle controls, and monitoring.
Eliminating MFA bypass requires architectural decisions, not just stricter prompts.
Map every way identities authenticate and authorize—human, service, legacy, automated, and third-party. MFA enforcement should be evaluated per path.
Session lifetime, token refresh behavior, and revocation processes matter as much as initial authentication.
Treat app consent as privileged access. Monitor for over-scoped grants, unusual API usage, and persistence through applications—not users.
Build detections around improbable success patterns: new apps, unusual token use, access without interactive logins.