Adversary-in-the-middle (AiTM) phishing is an attack where a proxy sits between the victim and the real login page, relaying the session so the attacker captures the authenticated session token, defeating most forms of multi-factor authentication. Unlike classic phishing that steals a password, AiTM steals the logged-in session itself.
Key facts
- The victim sees the genuine service (because the proxy relays it), enters credentials and completes MFA as normal; the attacker harvests the resulting session cookie.
- A stolen session token means the attacker is already authenticated: no password re-entry, no MFA prompt.
- Off-the-shelf AiTM phishing kits made this mainstream, and it is a leading route to account takeover and business email compromise.
Why AiTM defeats standard MFA
Most MFA proves who is logging in at the moment of login. AiTM lets that legitimate login happen, then rides the result: the session cookie issued after successful MFA is what gets stolen and replayed. The authentication succeeded; the session was hijacked. This is why "we have MFA" is no longer sufficient assurance against phishing, and why the defensive frontier moved to phishing-resistant methods and post-login monitoring. This page explains the concept; it does not provide attack tooling or steps. See also MFA fatigue and consent phishing for adjacent MFA-bypass routes.
How to defend
Phishing-resistant authentication (FIDO2/passkeys) is the strongest control, because it binds the login to the real domain and cannot be relayed by a proxy. Layer on conditional access (device compliance, location), token protection features where available, and behavioral monitoring for the tell that matters after the fact: a session suddenly used from new geography or device. On the mail side, blocking the AiTM lure before the click is the cheapest win, which is where email security carries the load.
How Sentaro helps
AiTM starts with a phishing message carrying the proxy link. Sentaro''s Message Defense evaluates that link''s infrastructure and intent before the victim reaches the proxy, and its Behavioral Defense flags the post-takeover mailbox behavior (anomalous sends, new rules) when a session is compromised despite defenses.