The rise of sophisticated social engineering and identity-based attacks has forced a re-evaluation of how organizations verify the legitimacy of internal and external requests. Recent incidents, ranging from high-stakes financial fraud to large-scale network breaches, highlight a common vulnerability: the reliance on human trust rather than robust, automated verification protocols. Whether through the use of generative AI or the exploitation of routine helpdesk procedures, attackers are increasingly bypassing technical defenses by manipulating the people tasked with maintaining them.
In January 2024, a finance employee at the Hong Kong office of engineering firm Arup was defrauded of approximately US$25.6 million. The incident began with an email appearing to originate from the company’s UK-based CFO. The employee, initially skeptical, was persuaded to authorize multiple transactions after participating in a video conference that featured deepfake versions of the CFO and other colleagues. Hong Kong police confirmed that the attackers utilized publicly available audio and video to create a pre-recorded, highly convincing simulation of a professional meeting.
A separate, large-scale breach occurred in April 2025 at the retailer Marks & Spencer (M&S). The company confirmed that attackers gained unauthorized access to their network by deceiving employees at a third-party contractor, rather than by exploiting a technical flaw within M&S systems. The resulting disruption lasted for weeks, forcing the suspension of online orders and affecting essential services like contactless payments and Click & Collect. M&S projected that the incident would reduce its 2025/26 operating profit by approximately 300 million. The attack has been attributed to the Scattered Spider cybercriminal group, a collective known for leveraging social engineering to penetrate corporate environments.
In another trend, Microsoft has reported a campaign where threat actors impersonate internal IT support staff via Microsoft Teams. By contacting employees through external messages or calls, attackers create a sense of urgency, pressuring staff to grant remote control of their devices under the guise of resolving technical issues. Once access is granted through tools like Quick Assist or screen-sharing software, the attackers use PowerShell to deploy malicious tools, establish persistence, and conduct reconnaissance. Microsoft observed these actors attempting to move laterally toward sensitive infrastructure, including domain controllers and certificate authorities.
These incidents underscore a critical shift in the threat landscape: attackers no longer require complex technical exploits when they can simply impersonate trusted figures. By gathering intelligence on an organization’s hierarchy and workflows, threat actors can craft pretexts that mirror routine business operations. This is particularly effective when targeting helpdesks or third-party contractors, where the pressure to provide rapid service can override standard security verification steps.
The reliance on traditional, knowledge-based verification—such as security questions or personal information—is increasingly insufficient, as this data is often easily accessible to attackers. Security experts emphasize that organizations must move toward multi-factor authentication (MFA) methods that do not rely on easily spoofed information. Integrating identity verification directly into service desk workflows and internal applications, such as chatbots or automated password reset portals, is considered a necessary step to remove the burden of verification from individual employees.
Beyond automated checks, organizations are being advised to implement independent confirmation channels for high-risk actions, such as large financial transfers or the granting of privileged access. By requiring verification through a secondary, pre-authorized path, companies can mitigate the risk of a single compromised communication channel. Furthermore, applying the principle of least privilege and maintaining rigorous monitoring for suspicious account activity remains essential to containing attackers who manage to gain an initial foothold.
Ultimately, the consensus among security professionals is that human awareness training, while valuable, cannot be the primary defense against modern impersonation tactics. As deepfake technology and social engineering become more refined, the focus must shift toward building verification into the fabric of business workflows. By ensuring that every sensitive request—whether for a password reset or a multi-million dollar transfer—undergoes a standardized, secure authentication process, organizations can better protect themselves against the erosion of trust that characterizes these modern cyber threats.




