The AI In-Lab Assistant is an intelligent chatbot integrated directly into the lab environment. It’s designed to provide on-demand help, guidance, hints, and tips for lab questions and tasks.
Its primary purpose is to enhance your learning experience by offering timely assistance without giving away direct answers, empowering you to solve challenges independently.
Contents:
- How to access and interact with the assistant
- What it can help with
- Availability and limitations
- Technology and data handling
- Data privacy and security
- Feedback and support
- Conversation history
- Guardrails
How to access and interact with the assistant
When you open a lab, you'll find the sparkle icon in the top right corner of your screen.
Clicking this icon will open the chatbot interface.
You can hide the chatbot by clicking the cross or the sparkle icon again. If the chatbot window obstructs your view, you can reposition it anywhere on the page by dragging its top bar.
What it can help with
The chatbot is designed to assist with lab-related queries and tasks. You can ask it to:
- Summarize the lab briefing
- Help you understand lab concepts (e.g., "what is a SIEM?")
- Provide assistance with specific tasks within a lab (e.g., "I need help with Task 3")
The chatbot will offer hints and tips to guide you toward solutions, but it won’t provide direct answers to questions or tasks.
For example, if you ask for the answer to Question 3, it will respond with guidance on how to figure it out yourself. It also can’t provide raw lab source material.
Availability and limitations
Like other AI features on the platform, the AI In-Lab Assistant can be accessed by enabling the generative AI function within platform settings. However, it won’t be available for organizations that don’t have generative AI enabled.
The AI In-Lab Assistant can be disabled by toggling the Gen AI switch to the OFF position within platform settings. (Please note this will turn off all Gen AI features, including AI Lab Builder and AI Crisis Sim builder).
The AI In-Lab Assistant is available in most labs, with the following exceptions:
Assessments: To ensure the integrity of your results, the assistant is disabled in Adaptive Assessments and Demonstrate Labs.
Custom Content: To protect proprietary information, the assistant is excluded from all Custom Content Labs.
A Note on Licensed Content & Privacy You may notice the chatbot interface appearing in certain labs protected by specific organizational licenses. Please be assured that your proprietary content is not being ingested by the AI service. While the bot may be visible, we have strictly blocked it from accessing the context or data within these licensed labs. Because the AI cannot "see" your specific lab content, it will be unable to provide assistance or summarize briefing panels in these environments.
Technology and data handling
Our AI service is built using OpenAI technology. It doesn’t have access to any customer data.
Data privacy and security
We prioritize your data security. The AI In-Lab Assistant only stores your account UUID in its service database. We don’t store any personally identifiable information (PII) other than what you explicitly input as queries, which OpenAI will also receive. We don’t ingest any custom content labs into our AI service, and the service doesn’t have access to any customer data.
Feedback and support
You can provide feedback on the chatbot's responses using the thumbs-up and thumbs-down icons located directly under each response. This feedback helps us monitor and improve the AI In-Lab Assistant over time.
If you’re unsatisfied with a response, you can also reach out to Customer Support through the usual channels. For any technical issues with the AI In-Lab Assistant feature itself, please raise this with Customer Support.
Conversation history
Your chat history will be lost when you exit a lab or move to a new one. To start a new chat in a lab, click the '+' icon next to "New chat" in the top left of the chatbot window to clear your current chat history.
Guardrails
We utilize OpenAI’s Moderation endpoint. We allow the illicit category for OpenAI, which covers topics such as hacking.
Comments
0 comments
Article is closed for comments.