SecurityNews

More than half of employees do not always verify the sender of sensitive requests

Nearly half of employees trust digital messages more than they did five years ago. At the same time, 55 per cent do not always check whether a request for money, login credentials or confidential information genuinely comes from the stated sender. That is the finding of research commissioned by WatchGuard Technologies among more than 500 full-time employees in eight countries.

The research, published as the Cyber Trust Index, maps which signals employees rely on when assessing a digital request. Familiar characteristics turn out to carry considerable weight. Of those surveyed, 72 per cent have at some point clicked a link or opened an attachment because it appeared to come from someone they knew.

Familiar signals carry weight

For 38 per cent, a familiar sender makes a digital request trustworthy. In addition, 36 per cent find a request credible if it contains familiar information about their work or private life, and 32 per cent are convinced by correct company logos or branding. Independent verification is cited by 27 per cent as a reason to trust a request. WatchGuard points out that the characteristics mentioned are becoming increasingly easy to imitate with AI.

Requests on behalf of management

Messages that appear to come from senior management are also often acted upon without verification. Of employees, 42 per cent would probably trust an urgent text message from their director or manager requesting payment or sensitive information without checking the sender. For a comparable voice message, the figure is 40 per cent. In the United States, 59 per cent would trust the text message.

When faced with an unusual or sensitive video request, 53 per cent rely on the face, the voice or the displayed name to establish the sender’s identity. Only 28 per cent opt for a different communication channel or an internal procedure. Asked about the biggest obstacle to being properly prepared for cyber threats, 38 per cent cite insufficient training or awareness among staff.

Differences between countries and sectors

How often employees check the sender varies widely by country. In Australia, 69 per cent do so, compared with 22 per cent in Germany. Among sectors, healthcare scores highest, with 63 per cent of staff verifying requests. Employees in financial services have the most trust in digital content. Public sector employees are the most sceptical about digital communication and text messages from managers, but even in that group fewer than half verify sensitive requests.

A role for MSPs

According to WatchGuard, MSPs can support organisations in setting up checks on sensitive requests. As an example, the company cites documenting how staff confirm payment requests, password resets and requests for confidential data via a second channel. It also recommends practising with realistic scenarios in which a manager is impersonated, and agreeing on permitted AI tools, data processing and human oversight.

WatchGuard also advocates easily accessible reporting channels and layered security. Alongside technical security, organisations should also measure whether staff verify and report suspicious requests. In this context, WatchGuard refers to its own solutions for identity security and secure access, in which users and their access rights are continuously verified, supplemented by Managed Detection and Response (MDR). The company claims these services reduce account misuse and AI-enabled fraud.

Familiar characteristics offer no certainty

Corey Nachreiner, Chief Information Security Officer at WatchGuard, says that familiar names, voices, faces and brands help people quickly judge whether they can trust a message. According to him, AI is making it ever easier to imitate those characteristics, meaning they no longer provide certainty about someone’s identity. Nachreiner expects verifying the sender to become one of the most important skills in cybersecurity as a result.

The full report can be read via the Cyber Trust Index.

This article was originally published in Dutch on MSP Business. Read the original →