All use cases

Social engineering scams

Catch scams as they happen by catching the moments fraudsters coach victims through high risk actions.

Overview

Social engineering scams work by manipulating a real user into acting against their own interest, usually while a fraudster coaches them in real time over the phone. The caller impersonates a bank, a platform's support team, the police, or a supplier, manufactures urgency, and talks the victim through reading out a one-time passcode, approving a transfer, resetting a password, raising a withdrawal limit, or buying gift cards. Every control built to catch an intruder fails here: the credentials are real, authentication succeeds, the device is the customer's own and already recognized, the location is normal, and no automation is involved.

The defenses businesses reach for first, customer education, warning interstitials, transaction limits, and callbacks, all depend on the victim recognizing a scam they are actively being talked out of recognizing. Fingerprint's Active Call Detection adds the missing piece by reporting whether the device has a cellular or internet (VoIP) call in progress at the moment of identification. Rather than inferring manipulation from behavior, it observes the coaching channel directly, and it does so without any risky or sensitive permissions and without collecting PII: no phone numbers, no call logs, no duration, nothing about the other party. Developer Tools Detection covers the adjacent remote-access variant, where the victim is talked into installing screen-sharing or support software.

The critical design constraint is that being on a call is completely normal. Plenty of people talk to a friend, a colleague, or a real support agent while using an app, so an active call is only meaningful in combination with what the user is doing at that moment. The pattern worth acting on is a call that coincides with a specific high-stakes step: passcode entry, a transfer to a new recipient, a credential reset, a phone number or withdrawal limit change, or a bulk gift card purchase. Scams concentrate in banking and fintech, where vishing losses are largest, with real exposure across e-commerce, gambling, gaming, and SaaS wherever stored value or account control can be handed over. Respond with graduated friction, an in-app warning naming the scam pattern, a cooling-off delay, step-up verification, or routing to manual review, rather than a hard block that punishes everyone on a legitimate call.

How to implement it

  • Check Active Call Detection at one-time passcode entry, payment and transfer approval, login and account recovery, high-risk settings changes to detect potential scam coaching.

  • Respond with graduated friction: an in-app warning that names the scam pattern ("Is someone on the phone asking you to do this?"), a cooling-off delay, step-up verification through a separate channel, or routing to manual review.

  • Use Developer Tools Detection to catch the remote-access variant, where the victim is coached into installing screen-sharing or support software instead of, or as well as, staying on the call.

Also known as

social engineering · social engineering scam · vishing · voice phishing · phone scam · scam call · live coaching fraud · coached transaction · scam in progress · impersonation scam · bank impersonation scam · tech support scam · phantom hacker scam · remote access scam · OTP scam · one-time passcode scam · passcode phishing · gift card scam · romance scam · investment scam · elder fraud · active call detection · scam detection

Ready to identify?

Start building with a free trial or speak to an expert to explore our Enterprise plan.