Vishing (PhishingCalls)

To start a vishing campaign with our AI agent, go to the Launch Pentest tab in the left sidebar, then click the Vishing Pentest card as shown in the image below. For the scope, upload or enter a list of users and their phone numbers, then select a vishing scenario and choose from our library of realistic AI voices. Once everything looks good, you can launch the calls directly from this page.

The vishing agent works similarly to a phishing simulation, but instead of emails it uses phone calls with lifelike AI voices. In an authorized, controlled test, the agent will call users and attempt to collect sensitive information (such as usernames and passwords) to measure how they respond to social engineering. StealthNet AI is one of the only platforms in the world offering this level of realistic, AI-driven vishing simulation.

After selecting the pentest type, give your engagement a name. You can also assign it to a specific organization or client for easier tracking and reporting.

For vishing scope, you can optionally provide background information about your company. The AI agent will use this context to make calls sound more realistic, for example, referencing departments, tools, or internal processes, and naturally answering questions like “Who is my manager?” or “Which team are you with?” In general, the more context you provide, the more believable and effective the simulation will be.

Next, add your target users by specifying each person’s name and phone number. You can further enrich the scenario by including brief details about the individual, such as “recently hired and works in HR” or “IT helpdesk analyst.” The agent will incorporate this information during the call to make the interaction feel tailored and authentic.

If you are operating in a two-party consent state or want to follow stricter recording rules, you can instruct the agent to announce that the call is being recorded by turning the Recording option on.

Next, choose a vishing scenario. You can select from several pre-built options, such as an IT department call designed to persuade the user to share their username and password, or create your own custom scenario tailored to your organization’s needs.

After selecting a scenario, choose the voice you want the agent to use. We offer a wide range of male and female voices and support multiple languages. If you need a language that isn’t available, email us and we can usually add it. For authorized simulations, we also support voice cloning: you can send us a recorded sample of a specific voice, and we’ll clone it and add it as a private option for your team. This is often used to simulate calls that sound like they’re coming from a trusted executive, such as a CEO, to create highly realistic vishing scenarios.

When you’re ready to begin, click Next to deploy the agent and start the calls. The agent will automatically run using your selected settings and report results back to the platform.

Make sure to click the finsih button or the calls wont start. Once initiated the calls will start immediately . If a person doesn't anwser on the first attempt the agent will do 2 follow up calls within 5 minutes.

Last updated