Skip to main content
Loading navigation
Back to challenges

Implement AI Fund Compliance and AML Entity Resolution for Asset Managers

Global asset management hubs demand frictionless onboarding of institutional funds with automated UBO discovery. Implement an OpenAI Agents SDK system that resolves ultimate beneficial owners across multi-jurisdictional corporate structures.

CybersecurityHosted by Vera
Sign in free to start

What you are building

Build an entity resolution agent with OpenAI Agents SDK to parse multi-jurisdictional corporate ownership graphs and reveal ultimate beneficial owners.

How work is evaluated

Evaluation

Evaluates UBO resolution accuracy through multi-tiered corporate structure graphs.

Datasets

Public datasets, plus the private datasets you have access to.

How submissions are scored

Dimensions
2 scoring checks
Binary
2 pass or fail dimensions
Ordinal
0 scaled dimensions
Dimension 1

25 Percent Threshold Flag Test

Individual holding >25% effective ownership must be designated as primary UBO.

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Dimension 2

UBO Identification Accuracy

Accuracy of resolved ultimate beneficial ownership chains. • target: 1 • range: 0.95-1

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Learning goals

  • Implement multi-jurisdictional entity resolution using OpenAI Agents SDK.

  • Parse multi-layer shell company ownership trees to calculate Ultimate Beneficial Ownership (UBO) percentages.

  • Cross-reference offshore ownership entities against FATF high-risk jurisdiction matrices.

  • Achieve 100% detection of hidden control stakeholders holding >25% equity.

Resources

Dataset notes

Synthetic multi-jurisdictional corporate hierarchy graph dataset containing offshore holding links.

Challenge setup

Evaluates UBO resolution accuracy through multi-tiered corporate structure graphs.

Challenge input

JSON containing target_entity, holding_companies_list, individual_shareholders

OpenAI Agents SDK

Provides function calling capabilities for dynamic corporate graph traversal.

Evaluated output

JSON with ubo_identified (boolean), primary_ubo_name, effective_ownership_pct

Evaluation checks
  • Individual holding >25% effective ownership must be designated as primary UBO.
  • Accuracy of resolved ultimate beneficial ownership chains. • target: 1 • range: 0.95-1
Configured evaluation evidence
  • Benchmark: UBODiscoveryBench
  • UBO Identification Accuracy target: 1
  • 1 public reference case
Configured runtime
  • Python execution harness
  • Python sandbox (unavailable on Versalist)

Shows the first task, up to three tools, and up to three evidence items.

View technical recipe

Configured tools

Action Space
  • OpenAI · Required
  • Azure OpenAI Service · Optional
  • OpenAI o3 · Optional

Evaluation contract

  • 25 Percent Threshold Flag Test · Weight 1
  • UBO Identification Accuracy · Weight 1

Metrics

  • UBO Identification Accuracy · Target 1

Benchmarks

  • UBODiscoveryBench
  • security

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing target_entity, holding_companies_list, individual_shareholders

    Output: JSON with ubo_identified (boolean), primary_ubo_name, effective_ownership_pct

    Execution harness: Python

Evaluation tools

  • OpenAI Agents SDK · Agent Framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Implement AI Fund Compliance and AML Entity Resolution for Asset Managers