Reverse Social Engineering
Reverse social engineering is an attack in which the victim is manipulated into contacting the attacker for help — making the interaction feel self-initiated and therefore trustworthy.
Reverse social engineering inverts the usual direction of a social engineering attack. Instead of approaching the victim, the attacker engineers a situation in which the victim reaches out to them — for help, support, or information. Because the victim initiated the contact, the normal skepticism toward unsolicited callers never activates: people rarely question the legitimacy of someone they chose to call.
How it works
The classic model has three stages. First, sabotage or pretext: the attacker creates a problem, or the appearance of one — a disconnected service, a planted error message, a warning email that a mailbox is about to be suspended. Second, advertising: the attacker positions themselves as the solution, by planting a "support" phone number in the error message, posing beforehand as the new IT contact, or listing a fraudulent helpline where victims will search for one. Third, assisting: the victim calls, the attacker helpfully "fixes" the problem — and in the process harvests credentials, has the victim install remote-access software, or extracts sensitive information.
Modern variants are everywhere once you know the shape: fake invoice emails whose real payload is the billing-dispute phone number (callback phishing), search ads for fraudulent airline or bank support lines, and browser pop-ups telling the victim to call "Microsoft." The same self-initiated-trust dynamic powers tech support scams at consumer scale, and in the enterprise it mirrors the trust confusion exploited in help desk impersonation attacks.
How to defend against it
- Publish one canonical support channel and train employees that IT will never be reached through numbers embedded in error messages, emails, or search ads.
- Treat inbound "I was told to call you" requests with the same verification as outbound ones — the direction of contact proves nothing about identity.
- Simulate it. Include planted-number and callback lures in awareness exercises, not just classic phishing links.
- Make verification cheap. A searchable internal directory and a norm of hanging up and calling back through a known number removes the attacker's main advantage.
Related terms
NOUSEC simulates attacks across 8 channels and turns the results into one number your board can read.
Book a demo