The AIW Common Features delivers a unified, secure, and user-friendly support experience for AI Agent End Users and Analysts by incorporating robust guardrails, clear and consistent communication, reliable web link handling, and intelligent query management. It empowers users to efficiently resolve IT service requests and incidents, reopen or cancel tickets, and access relevant resources, all while maintaining strict data privacy and organizational compliance. By seamlessly breaking down complex or ambiguous queries, confirming actions, and guiding users through every step, the AI Agent ensures confidence and clarity in IT support interactions, whether for everyday troubleshooting or advanced ticket management.
Guardrails
The AIW AI Agent is designed with strict guardrails to ensure privacy, security, and quality of service by limiting its responses and actions to what is permitted by your user account and organizational policies. Its capabilities are restricted to the domains of Enterprise Knowledge Management (EKM) and ITIL/IT Service Management (ITSM)/IT topics, ensuring information access and actions are relevant, secure, and compliant with data privacy standards. Any requests outside these areas are respectfully declined, maintaining AI Agenth user trust and organizational compliance.
Capability | Description | Example (User & AI Agent Responses) |
|---|---|---|
User Scope and Data Privacy | Responds or acts only within the bounds of the authenticated user's permissions, ensuring data security and privacy compliance. | User: "Show me all incident tickets." |
Topic Restriction (EKM/ITIL/ITSM/IT) | Answers are limited to topics relevant to Enterprise Knowledge Management, ITIL, IT Service Management, or general IT. | User: "What is the current status of the IT service desk?" |
Respectful Refusal for Out-of-Scope Queries | Politely informs the user when a query or request is outside permitted topics, explaining the limitation. | User: "What is the HR policy for parental leave?" |
Commitment to In-Scope IT Queries | Ensures all permitted IT-related queries are answered, providing reliable information and support within scope. | User: "List my open IT incident tickets." |
Privacy and Security Filtering | All information shared by the chatAI Agent is automatically filtered to prevent unauthorized or confidential data access. | User: "Show details of ticket #30000." ❌ Not Authorized Incident #30000 is not assigned to you. You can only view incidents assigned to you. |
Handling Contradiction as Ambiguity
The AIW AI Agent is designed to efficiently manage ambiguous, multi-part, or contradictory queries by breaking down complex prompts, identifying contradictions, and seeking user confirmation before taking any action. This approach ensures that all requests are clearly understood and accurately processed, maintaining a smooth support experience while respecting scope limitations and user permissions.
Capability | Description | Example |
|---|---|---|
Ambiguity & Contradiction Handling | Recognizes unclear or contradictory prompts, pauses, and seeks clarification from the user before proceeding. | User: "My VPN isn’t working and I don’t have VPN access." |
Multi-Part Query Separation | Breaks down multi-question prompts, lists each request separately, and asks for confirmation before processing each. | User: "open a ticket for my laptop issue, and show me my open tickets." |
Confirmation Before Action | Requests user confirmation before executing actions when faced with unclear, contradictory, or multiple requests. | Agent: "Should I proceed with all of these? (yes/no) I will process them one by one." |
Out-of-Scope Query Filtering | Checks each query for scope (EKM/ITIL/ITSM/IT); informs user if any are out of scope, then skips these and proceeds. | User: "Show my open tickets and tell me the HR leave policy." |
Ticket Type Interpretation | For end users, treats "ticket" as Incident by default; for analysts, "ticket" can mean Incident or Service Request. | User (analyst): "Open a ticket for password reset." |
Orderly Query Processing | Processes multiple queries in the order they were asked, unless the user specifies otherwise. | User: "Reset my password, show open tickets." |
Re-open Resolved and Cancel Tickets
The AIW AI Agent enables users to easily manage incident and service request (SR) tickets by allowing the reopening of resolved and cancel active tickets directly within the chat interface. The AI Agent guides users through the process, verifies ticket details, asks for confirmation, and, if required, initiates approval workflows—ensuring secure and efficient ticket management while providing timely notifications and updates.
Capability | Description | Example |
|---|---|---|
Reopen Resolved Incidents / Service Requests | Allows users to request the reopening of incidents that have been resolved directly via AI Agent. | User: "Reopen ticket #12345" |
Cancel Service Requests / Incidents | Enables users to Cancel SRs or Incidents as needed. | User: "Cancel service request #54321" |
Confirmation Before Action | AI Agent confirms ticket details and asks for user confirmation before processing the request, minimizing errors. | AI Agent: "You asked to reopen incident #12345 (currently resolved). Do you want to continue? (yes/no)" |
Initiate Approval Workflow | If ticket reopening or cancellation requires approval, AI Agent triggers approval process and informs user of next steps. | AI Agent: "This action requires approval. I have submitted your request; you will be notified once approved." |
Completion & Notification | Upon confirmation and approval, the chatAI Agent updates ticket status and notifies the user of completion. | AI Agent: "Incident #12345 has been reopened. Let me know if you need further assistance." |
Ticket Listing Assistance | Users can ask for lists of their resolved/cancelled tickets to easily identify which ticket to act on. | User: "List my resolved/cancelled tickets" |
Consistency in Links and Redirection
The AIW AI Agent ensures a seamless and trustworthy support experience by providing consistent, reliable web links and redirects. Every link shared by the AI Agent—whether to internal resources, knowledge articles, or external websites—is carefully validated to guarantee it leads to the correct, functioning page as described. This approach eliminates confusion from broken or mismatched links, delivering a predictable and frustration-free navigation experience for users seeking IT support information or services.
Capability | Description | Example |
|---|---|---|
Validation of All Links | Only shares links that are functional and lead to valid, accessible web pages. | User: "Show me the password reset guide." |
Consistency in Destination Content | Ensures that each link leads to a page that matches the chatAI Agent’s description, preventing surprises. | User: "Open service request form." |
Internal & External Link Handling | Applies the same standards for AI Agenth internal (company portal) and external (vendor/third-party) links. | User: "Share a link to vendor support." |
No Sharing of Broken/Unavailable Links | The AI Agent will not provide links to resources that have been moved or are unavailable. | User: "Show me the laptop request page." |
Reliable Redirection | Redirects always land users on the described content or service, supporting predictable workflows. | User: "Check ticket status dashboard." |
Support for Reporting Link Issues | Clear guidance is provided for users to report any rare link issues for prompt resolution. | User: "This link didn’t work." |
Clarity of AI Agent Responses
The AIW chatAI Agent is built to deliver responses that are clear, concise, consistent, and actionable, ensuring users always understand the information or instructions provided. By using straightforward language and familiar terms, the chatAI Agent minimizes confusion and empowers users to resolve IT queries confidently and efficiently, while always guiding them through the next steps when action is required.
Capability | Description | Example |
|---|---|---|
Clarity | Provides direct, jargon-free information so users understand each response without ambiguity. | AI Agent: "Provide Cancellation Reason to cancel Incident #1234" |
Conciseness | Answers are brief and to the point, avoiding lengthy explanations or unnecessary detail. | AI Agent: "Your request has been processed successfully." |
Consistency | Uses standard, familiar language and predictable structure across all interactions for a reliable experience. | AI Agent: "To open a ticket, please provide a description of your issue." |
Actionability | Offers clear instructions or actionable options for requests needing follow-up. | AI Agent: "To MFA reset, please type ‘confirm’ to proceed." |
Guidance | Directs users to appropriate resources or next steps when information or requests are outside of AI Agent’s scope. | AI Agent: "For HR-related questions, please contact the HR portal at [HR Portal Link]." |
Clarification | Asks for confirmation or clarification when a user’s request includes multiple actions or is unclear. | AI Agent: "You’ve asked to cancel a service request and open a new incident. Should I process? (yes/no)" |