Attackers target access, not authentication
Once authenticated, identities generate artifacts—tokens, sessions, delegated permissions—that can be reused without triggering MFA or interactive login events.
Defenders still think in terms of stolen passwords. Attackers don’t. Modern identity compromise increasingly bypasses credentials entirely, abusing session tokens, refresh tokens, OAuth grants, and delegated access to operate quietly inside cloud environments.
The identity perimeter moved. Most defenses didn’t.
Password theft is increasingly noisy, detectable, and mitigated by MFA. Token theft, by contrast, exploits the mechanics of modern identity systems: session lifetimes, delegated trust, and application permissions that were designed for convenience, not adversarial environments.
Once authenticated, identities generate artifacts—tokens, sessions, delegated permissions—that can be reused without triggering MFA or interactive login events.
Most SOC detections focus on failed logins, brute force, or phishing. Token abuse produces none of these signals.
Token-based access blends into normal cloud activity, extending dwell time and increasing impact before detection.
Password theft dominated the last decade of breaches. It still happens, but it increasingly fails to deliver reliable access.
Phished credentials without MFA access often lead to failed logins, push fatigue alerts, or blocked attempts that draw attention.
Improved password hygiene, managers, and identity providers have reduced the payoff of simple credential reuse.
Failed logins, geographic anomalies, and authentication errors create telemetry that defenders expect and monitor.
Tokens are designed to represent trust after authentication. Attackers abuse that trust directly.
MFA is checked during authentication, not every request. Stolen session or refresh tokens inherit prior trust.
API calls, app access, and session reuse appear identical to normal user or service behavior.
Long-lived sessions and refresh mechanisms turn a single compromise into sustained access.
Tokens issued to applications can access mail, files, and data at scale—often with less scrutiny than users.
Token theft rarely starts at the identity provider. It starts where trust is materialized.
Session cookies and tokens stored in browsers are harvested via malware, extensions, or local access.
Malicious or over-scoped apps receive tokens directly through legitimate consent mechanisms.
Non-human identities often lack strong monitoring, making token abuse harder to detect.
Token theft undermines governance assumptions about identity, control, and accountability.
High MFA adoption does not equal low identity risk if access persists through tokens and apps.
Most controls focus on authentication, not post-authentication access and session abuse.
Token abuse blurs the line between attacker and legitimate activity, complicating investigation and response.
Defending against token theft requires architectural change, not more password controls.
Treat session lifetime as a security control. Reauthentication and token revocation must be deliberate and tested.
App consent and delegated permissions should require review, monitoring, and periodic revalidation.
Build detections for unusual token use, app activity, and access patterns that succeed without interactive login.
Move reporting from “MFA coverage” to “time-to-revoke access” and “blast radius of compromised trust.”