Immersive leverages the OpenAI API to allow you to quickly create new Crisis Simulation scenarios from scratch. You can also opt to specify a briefing prompt to increase the relevance to your organization and/or upload PDF documents to give additional context. The AI Scenario Generator also lets you edit the briefing and preview the scenario before publication. Once created, these scenarios can be edited and published as normal to your organization's catalog.
Creating a Crisis Simulation Scenario with the AI Scenario Generator
To create a new Crisis Simulation scenario using the AI Scenario Generator:
- Navigate to Exercise > Crisis Simulations, and then click Create Scenario.
- From the Create your scenario options, click Create with AI.
The Create with AI window appears. - Specify the following information:
- Scenario Title
- Industry affected: The following options are available: Energy and Infrastructure, Financial Services, Government, Healthcare, Logistics and Supply Chain, Manufacturing, Other, and Technology.
- Type of attack (Attack vector): The following options are available: Data Breach, Denial of Service, Insider Threat, Malicious Code, Other, Phishing, Ransomware, Remote Working, Supply Chain Compromise, Targeted Attack, Unauthorised Access, Vulnerability Disclosure/Reporting, and Zero-Day Exploit.
-
Who is the threat? (Threat actor): The following options are available: Criminal Groups, Current Employees, Customers, Disgruntled/Former Employees, Human Error, Nation State, Other, Political/Social Activists, Red Hat Hackers, and Terrorist Groups.
- (Optional) You can specify a Briefing prompt to give additional information about the scenario you want to create to increase the relevance to your organization and the outcomes you are trying to achieve. For further details, see What should I write in the Briefing prompt field? below.
- (Optional) In the Reference file section, you can click Upload or drag and drop a PDF to add additional context. For further details, see What documents should I upload in the Reference file field? below.
- Click Generate briefing.
The scenario briefing appears. - You can either:
- Click Edit briefing if you want to change the scenario briefing.
or - Accept the briefing as-is.
- Click Edit briefing if you want to change the scenario briefing.
- Click Generate scenario.
- When your scenario is ready, the following message appears: Your scenario is ready for review. You can either:
- Click Preview Scenario to preview your scenario.
or - Click Continue editing in builder.
- Click Preview Scenario to preview your scenario.
Note: Refer to Creating Custom Scenarios for information on how to edit, review, and publish a Crisis Simulation scenario.
What type of content is created by the AI Scenario Generator?
The AI Scenario Generator creates a Crisis Simulation scenario. These generations are based on publicly available data related to crisis management and create generic Crisis Simulation scenarios.
If you choose to add additional context via the briefing prompt or reference file field, the AI will take this into account when generating your scenario.
If your organization has enabled Crisis Sim - Share custom scenarios for enhanced AI responses, your previous and future custom scenarios will be used to generate scenarios tailored to your particular context. The sections below explain more about Scenario Sharing functionality.
Before being published, review the AI tool's outputs to ensure they meet your requirements (see section below).
What should I write in the Briefing prompt field?
The Briefing prompt field is a space for you to give additional information about the scenario you want to create to increase the relevance to your organization and the outcomes you are trying to achieve. Effective prompting of AI tools is important to ensure you get the best results. Here are some suggestions to get the most out of the tool:
- Instruct the AI to adopt a specific persona or role: This can significantly influence the tone and style of its response. For example, “Act as a Crisis Incident Expert designing an ongoing crisis simulation exercise.”
- Offer background information: Give the AI relevant details about the subject matter or the situation. This helps it understand the nuances to generate more accurate responses.
- Include examples: Providing examples of the desired output can be highly effective. This allows the AI to learn from your preferences and produce similar results.
- Consider the audience: Specify the target audience for the AI output. This will help the AI tailor its language and level of detail appropriately.
For more information and examples, please see our guidance documentation on effective prompting.
What documents should I upload in the Reference file field?
The purpose of this field is to allow you to add additional context in the form of news articles or reports on cyber attacks that might be relevant to your industry, organizational context, or any content that you want to focus on for exercising purposes.
The Reference file field will only accept PDF files. If you add a file into this field, it will be sent as additional context to the AI to generate your scenario. It will not be stored by our third party AI vendor or used to train its models. Users are responsible for any documents that are uploaded to the AI.
You are responsible for the content of any documents you upload. This includes ensuring that you have all necessary rights and permissions to share the content, and that it does not contain any sensitive, confidential, or personal information that you are not authorized to disclose to third-party AI.
Will I need to add anything else to my scenario?
The AI Scenario Generator currently only generates text content, so you may want to add additional rich media (for example, images and videos) to your scenario.
We also recommend enabling, adding, or checking certain elements after the scenario has been generated while in “edit mode” to get the most out of your Crisis Simulation.
This includes:
- Checking that you are happy with the text formatting and narrative content
- Checking that the injects and options generated link logically
- Checking that you are satisfied with the role
- Enabling and adding response feedback or performance indicators
- If you want to capture ranked response data, you will need to select the ranked options setting and add a rank (great, good, weak, or okay) to each response option to suit your organization's preferred situational response
- Turning on response confidence or justifications
How does management of the AI Scenario Generator in Immersive work?
Organization admins can choose to use the AI Scenario Generator in a layered approach:
- Without Scenario Sharing enabled but with document upload and free-text interaction enabled. (default)
- With Scenario Sharing enabled and with document upload and free-text interaction enabled.
- No AI access at all
These settings can be updated in the AI section on the organization's Platform Settings page.
What does "without Scenario Sharing" mean?
This means that the AI will not use information from previous or future custom scenarios you have created when generating your scenario.
The AI will only use the title you give your scenario, the pre-filled drop-down filters (currently organizational sector, attack vector, and threat actor), any information entered in the Briefing prompt field, and any documents added in the Reference file section to create your scenario.
This means that the scenario is likely to be more "generic" and less relevant to your particular context, but you can still edit the final version to make it more relevant to you - just like with our catalog scenarios.
What does "Scenario Sharing" mean?
If your organization chooses to enable Scenario Sharing, Immersive will include relevant information from any previous custom scenarios that you have published in the context window for requests to our third-party AI vendor. This will make the scenario more relevant to your particular context. Further details on the context window and exactly what data will be shared can be found below.
Will my data be used for model training?
No. None of the data you choose to share will be used by the third party AI vendor for the purpose of training its models. Shared data will only ever be included in the temporary "context window" of generation and will not be stored by the third-party AI vendor.
What is a "context window"?
A "context window" is an extension of the query sent to an AI model. It exists only during the processing of the query and is not saved by any third parties.
Which third-party AI vendor does Immersive work with?
Immersive is using technology provided by OpenAI to support this feature. OpenAI’s business terms of use are available at https://openai.com/policies/business-terms/.
We will notify you in advance if we intend to engage alternative third-party AI vendors to support feature releases containing AI.
If I choose to enable Scenario Sharing for AI, exactly what will be shared?
When enabled, the AI Scenario Generator will include data from your previously published (and any future) custom scenarios in the "context window" of its request.
The data captured from your custom scenarios will include the following:
Scenarios:
- Title
- Description
- Industry Sector
- Attack Vector
- Threat Actor
Injects & Options:
- Title
- Description
The data shared will not include feedback, exercise information, reporting, account or organization information fields, or metadata.
The tool is not intended to process personal information. We recommend that you review your custom scenarios before choosing to enable Scenario Sharing to check whether they contain any sensitive or personal information and ensure its removal.
Immersive will endeavor to scan and de-identify fields for any personal information; however, we do not guarantee that any automated de-identification will be 100% effective.
Is Immersive compliant with the UK National Cyber Security Centre Guidelines for secure AI system development?
Yes. Immersive has implemented secure design and development procedures in line with the NCSC Guidelines document.
In which Geographic Region will my scenario information be processed?
OpenAI’s servers are located in the US. All requests made to this third party vendor will therefore be transferred to the US.
In which Geographic Region will my scenario information be stored?
All AI Service data will be stored in the Immersive US region. This means that if you enable Scenario Sharing, the scenario data specified above will be stored in the Immersive US region until the scenario sharing option is switched off. At this point, the data will be deleted.
Disclaimer
The outputs are generated by artificial intelligence and are not verified by Immersive for accuracy and do not represent Immersive’s views. We do not make any warranty or guarantee as to the accuracy, completeness, or reliability of any outputs.