Social engineering is the deliberate manipulation of people into disclosing information, bypassing a control, or performing an action for an attacker. It commonly uses phishing, voice or text messages, impersonation, pretexting, and physical access attempts. In a threat model, the attacker targets trust, urgency, authority, or helpfulness rather than exploiting software directly. Successful deception can expose credentials or personal data, authorize fraudulent payments, enable malware delivery, or provide an initial foothold for account or network compromise.
Effective defenses make sensitive requests independently verifiable and limit the damage of a mistake. Use phishing-resistant multi-factor authentication where practical, least-privilege access, and approval or call-back procedures for payments, password resets, and changes to account or banking details. Staff should have a simple way to report suspected messages without penalty; security teams can then investigate related accounts, messages, and login activity, revoke exposed credentials, and contain follow-on access. Awareness training helps people recognize pretexts, but should reinforce these technical and procedural controls rather than rely on vigilance alone.