Decisions under pressure can turn a fraud attempt into an incident
An invoice inside a real supplier thread. A voice on the phone that sounds like the CEO. A team using an AI tool nobody approved because the sanctioned one was too slow. Controls can fail at the human step, so we treat that step as part of the work — facilitated by a psychologist, with the technical content owned by our security lead.
When you need this service
- Someone tried to defraud you with a payment request or a fake email from a manager.
- Payment rules exist on paper but are not followed under pressure.
- You want management training that is practical, not a lecture.
The benefit for your organisation
- Leaders practise how to verify suspicious requests and decide under pressure.
- Payment and AI rules become workable in daily practice.
- Management has documented training it can show.
What we offer
Structured sessions led by a psychologist that make the human side of a security decision visible: who decides, under what pressure, with which shortcuts — and what a clear decision protocol looks like. The output is documented, readable by management and usable as evidence of management training. It is not therapy, not a test of individuals, and not a simulated attack.
Executive fraud-decision workshop
Module on the DORA (EU Digital Operational Resilience Act) / NIS2 (EU cybersecurity directive) readiness and evidence pack, or commissioned on its own. Intake, a 90-minute session for up to eight leaders, two defensive scenarios — a fake payment request inside an existing supplier thread and an executive-impersonation call — and a written decision protocol and action record.
Analysis of unapproved AI tool use (shadow AI)
Module on the AI use and governance review. Structured, voluntary interviews with six to eight staff on why unsanctioned tools are used — workload, status, fear of looking slow — feeding the acceptable-use wording and the governance evidence. Themes only; no role or department is attributed.
Quarterly refresh
A new scenario, a 60-minute rehearsal and an updated action record. Scheduled, 30-day cancellation, no on-call element.
- Request receivedAn e-mail in a real supplier thread asks to change the bank account.
- Independent verificationCall the supplier on a number you already hold — not one in the e-mail.
- Approve or escalateConfirmed: approve. Not confirmed or unsure: escalate to the named owner.
- RecordLog who checked what, and the outcome.
Who delivers it
Angel Brenishki
Managing director — Cyber Security, Governance, Risk and Compliance leadOwns the technical scenarios and control recommendations, connecting security leadership experience with practical verification and escalation procedures.
Diana Ginova
Partner — business psychology and cyberpsychologyExperience in psychology, psychotherapy, work with corporate clients and occupational burnout. Developing her practice into cyberpsychology.
How we work and what we do not do
The €900 workshop and the €700 shadow-AI analysis are add-on prices when attached to a project. The quarterly refresh is €900 per quarter — four quarters are €3 600 a year. A standalone workshop is quoted separately. We assess situations and decisions, never individual employees.
How it is delivered
- A psychologist leads the sessions; Dyasol’s security lead owns every technical statement, scenario and control recommendation and attends or is reachable throughout.
- Sessions produce a written decision protocol and an action record your management can file as documented training. We do not describe any session as compliance with a specific law.
- Interviews are voluntary and reports carry themes only. In a small team, removing names is not always enough, so we also check whether a description or quote could identify a participant. Raw notes are destroyed once the report is accepted, and the report is never personnel evidence.
What we will not do
- No phishing-susceptibility scores or profiles on named employees.
- No insider-risk or “disgruntlement” indicators on individuals.
- No design or operation of social-engineering attacks, and no manipulation content tailored to individual vulnerabilities.
- No clinical therapy, diagnosis or crisis counselling under the Dyasol brand.
- No emergency or out-of-hours availability.
- No interview or survey output that identifies an individual.
Cyberpsychology as a direction
Cyberpsychology — manipulation of executives, decision-making in incidents and human factors in AI adoption — is the direction in which this practice is developing. The formats shown here are offered with the stated limits; new formats are described as pilots until there is enough delivery evidence to make broader claims.
Looking at workload and sustainable practices rather than fraud? See Workplace wellbeing.
Sample output
Demonstration example — not a real client.
| Scenario | Decision | Rule |
|---|---|---|
| A call from “the boss” asking for an urgent transfer | Verify by calling back a known number | Second approval for any payment to a new account |
| An email from a “supplier” with new bank details | Confirm through the channel in the contract | No account change by email alone |
A useful exercise can follow a changed bank account from the email to the payment decision. The fraudulent invoice guide includes a verification card and an illustrative scenario.