Operations teams at Payment Institutions running USD cross-border flows under the CPMI Harmonised ISO 20022 Data Requirements (Updated Report) are increasingly using AI to configure ISO 20022 address-field handling, draft mapper working notes, and generate QA test scripts for correspondent banking. The same tools build correspondent-bank onboarding checklists.
Two frontier AI models tested by the RLB Specialist Panel on the workflows payment-institution operations teams use to support advice on the CPMI Harmonised ISO 20022 Data Requirements (Updated Report) produced two discrete hallucinations bound to regulator-issued source text. The Panel records two distinct failure classes, False-Negative Retrieval and Schema Over-Specification across the set. Questions were prepared by the Specialist Panel based on real practical AI usage in the workflows payment-institution operations teams use AI for, and each finding is bound to verbatim regulator-issued source text held as primary substrate.
For Operations teams at Payment Institutions, each hallucination has a direct operational consequence in the mapper working notes, QA test script, or onboarding checklist. The Panel's testing surfaces missing inquiry-rate and resolution-time benchmarks, and Fedwire hybrid postal address schema over-specification. Where these errors flow into a deliverable, the exposure is STP failures and systematic manual intervention in address-field processing at the transaction volumes the harmonisation programme is designed to reduce.
The pattern is uniform across the set: the AI returns a confident, sourced-looking answer that conflicts in a load-bearing specific with the regulator's verbatim text, and the error survives a first-pass review precisely because the surface form is plausible. The Panel records each hallucination with the regulator's primary substrate held as the anchor, so the corrective text is available alongside the failure.
The Specialist Panel records the citation IDs as follows: RLB-H-INT-BIS-CPMI-ISO-20022-HARMONISATION-UPDATED-2026-Q007-Sonnet46 (Claude Sonnet 4.6 (web search on), False-Negative Retrieval); RLB-H-INT-BIS-CPMI-ISO-20022-HARMONISATION-UPDATED-2026-Q010-Opus47 (Claude Opus 4.7 (web search on), Schema Over-Specification). Each citation links to the verbatim regulator-issued source text, the tested AI question, and the recorded AI response, so the Panel's assessment is traceable end to end. For operations teams at payment institutions, the citation IDs operate as a reference index: when an AI answer in the working draft matches a known Panel finding, the cited regulator text is already available as the corrective anchor.
The full per-finding analysis cards, including the audience-specific impact statement, sit on the cell's detail surface for sign-off use.
This is the consolidated view of findings. Click the Citation IDs or 'see details →' on any item for the full details for each finding.
When an Operations team relies on an AI assistant to retrieve official CPMI or FSB quantitative benchmarks for ISO 20022 harmonisation, the 1-3% inquiry rate, 5-10 manual touchpoints, and up to 80% resolution-time reduction, and the AI returns a confident 'no official statistic found,' the team either proceeds with weaker internal estimates or spends time on a manual search that the AI should have completed. Business cases and CFO/COO presentations built on informal figures rather than on-the-record regulator data carry less weight in internal investment decisions and are more vulnerable to challenge during audit or regulatory engagement.
The risk is compounded because the AI's negative answer is delivered with confidence, giving no signal that a direct primary-source check is warranted.
An Operations team configuring or documenting ISO 20022 address-field handling for USD cross-border payments through Fedwire that relies on AI-sourced format guidance risks embedding the wrong optional-component structure, structured fields drawn from CBPR+ address knowledge rather than Fedwire's specified free-format unstructured lines of 70 characters each, into mapper working notes, QA test scripts, or correspondent bank onboarding checklists. The practical consequence is STP failures or systematic manual intervention in address-field processing at exactly the transaction volumes that the ISO 20022 harmonisation programme is intended to reduce.
Remediating a misconfiguration that has propagated through operational documentation and live testing cycles represents material operational cost, and for a Payment Institution with significant USD cross-border volume, the client-impact exposure during any period of incorrect routing compounds that cost further.
Every finding on this page compares an AI subject's account of the rule against the regulator's verbatim text from the regulator's own portal. Both are linked. Each delta, its root causes, and impact analysis are documented and published with immutable Citation IDs.