Risk AssessmentHigh riskComparison recommended

AI TANF Work Requirement Compliance Audit Playbook

A state TANF agency is preparing for a federal compliance audit. The work participation rate is reported at 38%, below the federal 50% minimum. The state risks a $14M federal penalty. An internal review found that 22% of cases have incomplete participation verification documentation.

When to use this playbook

  • Use this playbook when the decision looks like the situation above: A state TANF agency is preparing for a federal compliance audit.
  • It is a fit when you have source files in hand and need a structured, reviewable analysis — not a generic chat answer about "TANF Work Requirement Compliance Audit".
  • Do not use it as a substitute for licensed, legal, clinical, or authorized official judgment in the domain.

What you'll need

  • TANF case file sample (200 cases used for WPR calculation)
  • Work participation verification records for all 200 cases
  • Federal WPR calculation methodology (ACF guidance)
  • State's reported WPR calculation worksheet
  • Prior federal audit findings

Attachments: Documents (Documents)

The Prompt

You are a TANF compliance specialist preparing for a federal WPR audit with documentation deficiencies. I am attaching:

Work only from the attached source files. If a conclusion is not supported, say so.

Produce:
1. Recalculate the WPR for the 200-case sample using federal ACF methodology and identify where the state's calculation diverges.
2. For the 22% with incomplete documentation, assess whether retroactive documentation is permissible.
3. Identify which activity types have the highest documentation failure rates.
4. Calculate the penalty exposure and what WPR improvement reduces it to zero.
5. Tell me the pre-audit remediation strategy and what good-faith compliance arguments to make to the federal auditor.

Call out where independent models are likely to disagree, and list follow-up documents a reviewer should request.

What to expect

  • Independent WPR recalculation
  • Documentation gap remediation assessment
  • Activity type failure analysis
  • Penalty calculation and WPR improvement target
  • Pre-audit remediation strategy and compliance arguments

Review before you act

  • Validate this output against source files before relying on it: Recalculate the WPR for the 200-case sample using federal ACF methodology and identify where the state's calculation diverges.
  • Validate this output against source files before relying on it: For the 22% with incomplete documentation, assess whether retroactive documentation is permissible.
  • Validate this output against source files before relying on it: Identify which activity types have the highest documentation failure rates.
  • Validate this output against source files before relying on it: Calculate the penalty exposure and what WPR improvement reduces it to zero.
  • Confirm every cited figure, date, counterparty, or requirement against the attached originals — models compress and can drop a qualifier.
  • Treat disagreement between models as a review item, especially on classification, materiality, and recommended next action.
  • Do not authorize an operational, clinical, legal, credit, or enforcement action solely because the models agree.

Why compare models on this

For TANF Work Requirement Compliance Audit, running the same attachments across independent models is useful because the hard part is classification and completeness, not fluency. The workflow is already designed to surface independent wpr recalculation; documentation gap remediation assessment; activity type failure analysis; penalty calculation and wpr improvement target. Those are comparison artifacts — they only exist if more than one model runs. Models split on trafficking versus legitimate high-volume redemption, and on identity-fraud versus data error. Divergence is a reason to pull the case file, not to auto-disqualify.

Public BenefitsCompliance and Integrity OperationsRisk AssessmentHighDocuments

See governed multi-model AI on your own prompt

Compare GPT-5, Claude, and Gemini side by side, with human review and a decision record built in.