Appearance
Financial Use Cases
Real-world scenarios for financial services organizations using GhostAI.
Customer Service
Account Inquiries
Scenario: Customer calls about suspicious activity.
Agent Input:
Customer Mary Johnson, account 1234567890,
is reporting unauthorized charges on card ending 4242.
Transactions on 3/15: $500 at Amazon, $300 at Best Buy.
Please help me draft a dispute form.What AI Sees:
Customer [PERSON_a1b2], account [ACCOUNT_c3d4],
is reporting unauthorized charges on card ending [CARD_e5f6].
Transactions on [DATE_g7h8]: $500 at Amazon, $300 at Best Buy.
Please help me draft a dispute form.Result: Professional dispute form with customer details restored.
Credit Card Applications
Scenario: Processing new card application.
Input:
Review this credit application:
Applicant: James Wilson, SSN: 123-45-6789
Income: $75,000, Employer: Tech Corp
Credit score: 680, DTI: 28%Protected: All PII tokenized during AI analysis.
Lending Operations
Loan Underwriting
Scenario: Underwriter needs AI assistance with complex loan.
Input:
Evaluate this commercial loan:
Borrower: ABC Corp, EIN: 12-3456789
Contact: John Smith (CEO), SSN: 987-65-4321
Requested: $500,000 line of credit
Collateral: Equipment valued at $750,000Result: AI provides analysis with protected business and personal identifiers.
Collections
Scenario: Collections specialist drafting outreach.
Input:
Draft a collections letter for:
Account holder: Sarah Chen
Account: 5678901234, 60 days past due
Amount owed: $2,500
Payment options available: Settlement at 80%Protected: Account information secured during AI processing.
Wealth Management
Portfolio Review
Scenario: Advisor preparing client review.
Input:
Prepare portfolio summary for:
Client: Robert Brown, Account: WM-12345
Current holdings: See attached
Performance YTD: +8.5%
Upcoming RMD: $15,000 by 12/31Result: Personalized portfolio summary with protected client data.
Financial Planning
Scenario: Creating retirement projection.
Input:
Create retirement analysis for:
Client: Lisa Park, DOB: 06/15/1970
Current assets: $1.2M
Target retirement: 2035
Social Security estimate: $2,800/moProtected: All personal and financial data tokenized.
Compliance & Risk
SAR Preparation
Scenario: Preparing Suspicious Activity Report.
Input:
Help draft SAR narrative for:
Subject: John Doe, SSN: 111-22-3333
Account: 9876543210
Suspicious activity: Multiple large cash deposits
Pattern: $9,000 deposits on consecutive daysProtected: Subject information secured; AI helps with narrative structure.
Audit Support
Scenario: Preparing audit response.
Input:
Summarize account activity for audit:
Customer: ABC Corp
Account: 1234567890
Period: Q1 2024
Focus: Wire transfers over $10,000Result: Audit summary with protected account information.
Operations
Payment Processing
Scenario: Investigating failed payment.
Input:
Troubleshoot this payment failure:
Sender: Tech Corp, Account: 111222333
Receiver: Acme Inc, Routing: 021000021
Amount: $50,000, Error: Invalid accountProtected: All account and routing numbers tokenized.
Account Maintenance
Scenario: Processing account change request.
Input:
Process address change for:
Customer: Emily Davis
Account: 5555666677778888
Old address: 123 Old St, City, ST 12345
New address: 456 New Ave, Town, ST 67890Result: Change processed with protected customer information.
Best Practices
| Scenario | Data Types | Protection Level |
|---|---|---|
| Customer service | Name, account, card | Full |
| Lending | SSN, income, credit | Full |
| Wealth management | Account, holdings | Full |
| Compliance | All PII | Full |
| Operations | Account, routing | Full |
Key Takeaways
- Always use financial vertical: Enables specialized detection
- Review before sending: Check what's being tokenized
- Minimize card data: Use last-4 when possible
- Log everything: Maintain audit trail
- Train continuously: Keep staff aware of protections
Prompt Enhancement
The magic wand adds regulatory awareness and financial terminology to your prompts.
Example: Transaction Analysis
You type:
analyze these transactionsEnhanced:
Analyze the following transaction data for patterns and anomalies. Identify
potential compliance concerns (SAR triggers, structuring patterns), calculate
key metrics (volume, velocity, counterparty concentration), and present
findings in a format suitable for a BSA/AML review with risk ratings.PII Protection: Account numbers, routing numbers, and customer identifiers are tokenized before enhancement.