WPA3 and SAE: How Modern Wi-Fi Stops Offline Dictionary Attacks

WPA3 (Wi-Fi Protected Access 3) replaces WPA2’s pre-shared key exchange with SAE — Simultaneous Authentication of Equals — a password-authenticated key exchange that makes offline dictionary attacks against the Wi-Fi passphrase infeasible, even when the passphrase itself is weak. Where WPA2’s four-way handshake leaked enough material for attackers to test password guesses offline at massive speed, SAE forces every single guess to be an active, on-air exchange with the network. That one design change is the headline security improvement of WPA3-Personal and the fact the SY0-701 exam tests most often about modern wireless security.

The problem WPA3 was built to fix

In WPA2-Personal, every device shares one passphrase, which is converted into a pre-shared key (PSK). When a client joins, it runs a four-way handshake with the access point to derive session keys from that PSK. The fatal flaw: the handshake messages contain values computed from the PSK. An attacker who captures one handshake — trivially provoked by sending a deauthentication frame so a client reconnects — walks away with a file they can attack offline. They feed candidate passphrases into cracking software (Hashcat, Aircrack-ng), compute what the handshake would look like for each candidate, and compare. GPUs (graphics processing units) test millions to billions of guesses per second, silently, with no further contact with the network. A human-memorable passphrase rarely survives.

Note what the attack does not require: no interaction beyond one captured handshake, no alerts on the network, no rate limiting anywhere. The dictionary attack happens entirely on the attacker’s hardware. That is what “offline” means, and it is the specific property SAE eliminates.

How SAE works

SAE, defined in IEEE 802.11 and based on the Dragonfly key exchange, replaces the PSK derivation with a cryptographic negotiation between two equal peers (hence the name — neither side is the “server”; both authenticate simultaneously).

The handshake has two phases. In the commit phase, each side combines the passphrase with elliptic-curve cryptography to produce a commitment — a value that depends on the password but reveals nothing useful about it, because deriving the password back out requires solving a hard discrete-logarithm-style problem. In the confirm phase, each side proves to the other that it derived the same shared secret, which only happens if both know the passphrase. From that shared secret, fresh session keys are derived.

Three properties fall out of this design:

• No offline-testable material. A captured SAE exchange gives an eavesdropper nothing they can check password guesses against. Each guess can only be tested by running a live handshake with the AP — one guess per exchange, over the air, visible, and throttleable. A weak passphrase is still a bad idea, but cracking it goes from billions of silent guesses per second to a noisy handful.
• Forward secrecy. Session keys are derived from ephemeral values generated per handshake, not from the passphrase alone. An attacker who records encrypted traffic today and learns the passphrase next month still cannot decrypt the recording. Under WPA2-PSK, they could.
• Downgrade and clogging resistance. WPA3 includes protections against forcing clients back to WPA2 in transition-mode networks, and SAE includes an anti-clogging mechanism so attackers can’t cheaply exhaust an AP with bogus handshakes.

WPA3 certification also mandates PMF (Protected Management Frames), which authenticates management traffic and blunts the deauthentication attacks that made WPA2 handshake capture so easy in the first place.

WPA3-Personal vs WPA3-Enterprise

SAE is specifically the Personal mode story. Enterprise mode solves authentication a different way entirely:

	WPA3-Personal	WPA3-Enterprise
Authentication	Shared passphrase via SAE	Per-user credentials via 802.1X/EAP against a RADIUS server
Who knows the secret	Everyone on the network	No shared secret; each user/device has its own credential
Key derivation	SAE handshake per client	Derived from the EAP exchange per session
Optional strength tier	—	192-bit mode using the CNSA (Commercial National Security Algorithm) suite for high-security environments
Revoking one user	Change the passphrase everywhere	Disable one account or certificate centrally

The distinctions to hold onto: Enterprise mode does not use SAE or a passphrase at all — authentication happens through 802.1X against an authentication server, with individual credentials that can be revoked one user at a time, and it offers an optional 192-bit cryptographic mode that Personal mode lacks. The strongest pairing is WPA3-Enterprise with certificate-based EAP-TLS authentication, which removes passwords from the picture entirely. Personal mode’s achievement is different: it keeps the convenience of one shared passphrase while stripping out the offline-cracking exposure that made shared passphrases dangerous.

WPA3 additionally introduced Enhanced Open (OWE — Opportunistic Wireless Encryption) for public hotspots: encryption without any authentication, so open-network traffic is no longer readable by every nearby sniffer.

How the SY0-701 exam tests this

• The feature-identification pattern. A stem asks which WPA3 capability protects a pre-shared passphrase from offline dictionary attacks — sometimes adding “even with a weak passphrase.” The answer is SAE. Distractors typically include PMF, AES-GCMP, WPS, and TKIP; only SAE addresses offline guessing.
• The WPA2-weakness contrast. A migration scenario asks which characteristic of the SAE handshake fixes the four-way handshake’s flaw. Credited reasoning: captured SAE traffic can’t be used to verify password guesses offline — every guess requires a fresh active exchange with the network (with forward secrecy as a companion benefit).
• The Personal-vs-Enterprise discrimination. Multi-select questions ask what characterizes Enterprise mode in contrast to Personal: authentication through 802.1X/EAP and a RADIUS server rather than a shared passphrase, per-user credentials, and the optional 192-bit CNSA mode. Wrong options claim Enterprise uses SAE or a stronger passphrase.
• The recommend-an-upgrade pattern. An audit finds WPA2-PSK with evidence of captured handshakes being cracked, and asks the best remediation. WPA3-Personal (SAE) beats answers like “longer passphrase” or “hide the SSID.”

Wireless security questions surface across several exam domains — the full SY0-701 study guide shows where this topic fits in a complete study plan, and SY0-701 practice sets drill the SAE-versus-distractor discriminations.

Quick reference

• SAE = Simultaneous Authentication of Equals, WPA3-Personal’s replacement for the WPA2 PSK four-way handshake; based on the Dragonfly exchange.
• WPA2’s handshake leaks offline-verifiable material; one captured handshake enables GPU dictionary attacks at massive guess rates.
• SAE makes every password guess an active over-the-air exchange — offline cracking is off the table, even for weak passphrases.
• SAE provides forward secrecy: learning the passphrase later doesn’t decrypt previously captured traffic.
• WPA3 mandates Protected Management Frames, blunting deauthentication attacks.
• WPA3-Enterprise uses 802.1X/EAP with RADIUS — no passphrase, no SAE — plus an optional 192-bit CNSA mode.
• Enhanced Open (OWE) encrypts open hotspot traffic without authentication.
• Strongest wireless stack: WPA3-Enterprise with EAP-TLS client certificates.