Authentication Factors & MFA — Explain factors of authentication and MFA/2FA best practices (conceptual)

For most of computing history, verifying a user's identity relied on a single mechanism: the password. From early time-sharing systems at MIT in the 1960s to the first commercial web applications of the late 1990s, a simple username-and-password pair was considered sufficient to control access. As networks expanded and adversaries became more sophisticated, the inadequacy of this single-factor approach became painfully evident through a series of high-profile breaches that exposed millions of credentials at a time. The concept of requiring multiple independent proofs of identity grew from a niche military requirement into a mainstream security practice adopted by enterprises, banks, and eventually everyday consumer services.

The trajectory from single passwords to multi-factor architectures raises a fundamental question: how do we formally categorize the evidence a system can demand from a claimant, and how do we combine those categories to achieve a target assurance level? The answer lies in understanding the distinct authentication factor types and the security guarantees that emerge when they are composed together.

At its core, authentication is the process by which a system verifies the identity claim made by a principal—a user, device, or service. It is distinct from identification (asserting who you are) and authorization (determining what you may do). Authentication factors are the independent categories of evidence that a claimant can present. The three classical factors, often summarized as something you know, something you have, and something you are, form the foundation of modern identity verification. Two additional categories—somewhere you are and something you do—are sometimes recognized in extended frameworks, though the classical triad remains the industry standard referenced in NIST SP 800-63B and similar guidelines.

The central principle visible in this diagram is that the MFA gate requires factors from distinct categories, not merely multiple instances of the same category. Requiring two passwords (both knowledge factors) is not MFA—it is a stronger single-factor authentication at best, because a phishing attack that captures one password can just as easily capture the second. The security gain of MFA comes from forcing an adversary to mount fundamentally different attacks simultaneously: phishing to obtain the password and physical theft to obtain the hardware token, for instance. This orthogonality of attack surfaces is the defining property that makes MFA far more resilient than any single-factor scheme, regardless of how complex that single factor may be.

To understand why MFA provides a quantifiable security improvement, it is useful to reason about independent compromise probabilities. If the probability that an adversary compromises a single authentication factor is P(F), and the factors are drawn from independent categories, then the probability of an adversary successfully compromising all factors in an n-factor scheme follows a multiplicative model. This simple probabilistic framing clarifies the magnitude of security gain MFA provides over single-factor authentication.

The TOTP and FIDO2 equations above represent two major families of possession-factor protocols. TOTP is a symmetric shared-secret scheme—both server and authenticator hold the same key K, which means a server-side database breach can expose the TOTP secrets for all users. FIDO2, by contrast, is an asymmetric public-key scheme where the server stores only the public key; even a complete server compromise reveals nothing useful to an attacker who needs the private key locked inside the hardware authenticator. This distinction is central to understanding the security hierarchy among second-factor methods, which we explore in Section 5.

Not all second factors are created equal. Within the possession factor category alone, the mechanisms differ dramatically in their resistance to common attack vectors. Understanding this hierarchy is essential for making sound architectural decisions when designing or evaluating an authentication system. The spectrum ranges from SMS-based codes (weakest) to FIDO2 hardware security keys (strongest), with several options in between.

Several insights emerge from this hierarchy. First, SMS-based OTP is widely regarded as the weakest second factor still in common use—NIST SP 800-63B explicitly marks it as a restricted authenticator due to vulnerabilities in the SS7 telephony signaling protocol and the prevalence of SIM-swap fraud. Second, TOTP authenticator apps (Google Authenticator, Authy, etc.) eliminate the telephony attack surface but remain vulnerable to real-time phishing proxies like Evilginx, which intercept both password and TOTP code simultaneously. Third, push notifications with number matching improve upon basic push (which was susceptible to MFA fatigue attacks, as demonstrated in the 2022 Uber breach) by requiring the user to confirm a random number displayed on the login screen. Finally, FIDO2 security keys provide the strongest guarantees because the authentication challenge is cryptographically bound to the requesting origin, making proxy-based phishing mathematically impossible without the private key.

Suppose you are the security architect for a mid-sized university's IT department. You need to design an MFA policy for three user populations: general students accessing email, faculty accessing the learning management system (LMS), and IT administrators accessing infrastructure consoles. The goal is to select appropriate authentication factors and second-factor methods for each population, balancing security requirements against usability and cost constraints.

The evolution of authentication does not stop at traditional MFA. Three advanced paradigms represent the frontier of identity and access management, each building upon the foundational concepts covered in this lesson. Understanding how MFA connects to these frameworks provides the conceptual bridge to more advanced coursework in security architecture.

The trajectory toward passwordless, zero-trust, continuous authentication represents the logical endpoint of the principles covered in this lesson. If the goal of MFA is to layer independent verification factors, then eliminating the weakest factor (the password) while strengthening the remaining ones and extending verification across the entire session duration achieves the maximum benefit. The FIDO Alliance's passkey initiative, supported by Apple, Google, and Microsoft, is actively working to make this vision a consumer-level reality. As a CS student entering the security field, understanding both the classical MFA framework and these emerging paradigms will be essential for designing resilient identity architectures.