ResearchCritical riskComparison recommended

AI Playbook for Ponzi-Style Cash Flow Mapping

A state securities regulator has frozen the accounts of an investment fund that raised $22M from 340 investors. Early investors were paid 'returns' from new investor capital. You have complete bank records for 6 years and need to reconstruct the flow of funds for the receiver.

When to use this playbook

  • Use this playbook when the decision looks like the situation above: A state securities regulator has frozen the accounts of an investment fund that raised $22M from 340 investors.
  • It is a fit when you have source files in hand and need a structured, reviewable analysis — not a generic chat answer about "Ponzi-Style Cash Flow Mapping".
  • Do not use it as a substitute for licensed, legal, clinical, or authorized official judgment in the domain.

What you'll need

  • 6 years of bank statements (all accounts)
  • Investor subscription and redemption records (340 investors)
  • Wire transfer logs
  • Operator personal account statements
  • Fund offering documents

Attachments: Documents (Documents)

The Prompt

You are a forensic accountant reconstructing cash flows for a securities fraud receivership. I am attaching:

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

Produce:
1. Trace all investor deposits and identify how much of each 'return' payment was sourced from new investor capital versus actual investment returns.
2. Identify the net winner and net loser investors: who received more than they invested, and who received less?
3. Calculate the total clawback pool from net-winner investors and assess collectability by investor.
4. Map all transfers from fund accounts to operator personal accounts and identify whether any have a legitimate business basis.
5. Tell me the priority waterfall for distributing recovered assets and what the recovery rate looks like for net-loser investors at current asset levels.

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

What to expect

  • Investor net position table (winners and losers)
  • Cash flow tracing by period
  • Clawback pool estimate with collectability flags
  • Operator diversion register
  • Receiver distribution waterfall

Review before you act

  • Validate this output against source files before relying on it: Trace all investor deposits and identify how much of each 'return' payment was sourced from new investor capital versus actual investment returns.
  • Validate this output against source files before relying on it: Identify the net winner and net loser investors: who received more than they invested, and who received less?.
  • Validate this output against source files before relying on it: Calculate the total clawback pool from net-winner investors and assess collectability by investor.
  • Validate this output against source files before relying on it: Map all transfers from fund accounts to operator personal accounts and identify whether any have a legitimate business basis.
  • 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 Ponzi-Style Cash Flow Mapping, 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 investor net position table (winners and losers); cash flow tracing by period; clawback pool estimate with collectability flags; operator diversion register. Those are comparison artifacts — they only exist if more than one model runs. Models often split on qualitative materiality, intent versus error, and whether a newly formed counterparty is a red flag or a legitimate intermediary. Those splits are the review queue — not noise.

Forensic AccountingInventory and Cash SchemesResearchCriticalDocuments

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.