انتقل للمحتوى الرئيسي
الإصدار: 1.0.0-beta

مساعد الذكاء الاصطناعي

The AI Companion is AIIA's conversational assistant available across all modules. It helps auditors draft content, search historical data, analyze documents, and answer questions about audit standards.

الوصولing the مساعد الذكاء الاصطناعي

Click the AI Assistant icon (💬) in the bottom-right corner of any page. The companion panel opens as a side drawer.

What You Can Do

1. Ask Questions

"What controls should I test for accounts payable?"
"Summarize the IIA IPPF requirements for workpaper documentation"
"What were the key findings from last year's IT audit?"

2. مسودة Content

"Draft a workpaper conclusion for my AP duplicate payment testing"
"Write an executive summary for this engagement"
"Generate a scoping memo for a procurement audit"

3. Analyze Data

"How many open findings do we have by severity?"
"Show me audit plan completion vs target"
"What's the average finding resolution time this year?"

4. Search History (RAG)

"Find all findings related to access management from the past 2 years"
"Show me previous audits of the Finance department"
"What recommendations did we make for similar issues?"

5. Get Suggestions

"Suggest risks for the HR department"
"Recommend test procedures for change management"
"What KRIs should I monitor for IT operations?"

Context Awareness

The AI Companion is context-aware — it knows:

  • What page you're on — suggestions are relevant to the current module
  • What engagement you're working on — references the engagement scope
  • Your role — responses are tailored to your permissions
  • Your organization — only accesses data you're authorized to see

Response الميزةs

Every AI response includes:

FeatureDescription
Source CitationsLinks to specific workpapers, findings, or documents
Confidence ScoreHow certain the AI is about the response
Apply ButtonOne-click to apply AI-generated content
CopyCopy response to clipboard
FeedbackThumbs up/down to improve AI quality

Governance

  • The AI cannot modify any record without your explicit Apply action
  • Every interaction is logged in the audit trail
  • The AI only accesses data within your RBAC permissions
  • Low-confidence responses are flagged with a warning
  • You can review all AI interactions in Audit Logs → AI Actions

نصيحةs for Best Results

  1. Be specific — "Draft a conclusion for WP-003 AP testing" > "write something"
  2. Provide context — "Based on 4 of 25 samples failing, draft a finding"
  3. Iterate — ask follow-up questions to refine the output
  4. Verify — always review AI output before applying
  5. Use citations — check the source references for accuracy