Engineering and data teams at payments-API SaaS firms land on d224 and d230 at the public API contract, the ISO 20022 schema evolution layer, the address-normalisation pipeline and the customer release-notes template. Two AI failures on this regulation hit those surfaces directly. Opus 4.7 returned a reconstructed stakeholder taxonomy against d224's 10 recommendations, and Sonnet 4.6 committed to a November 2026 ISO 20022 structured-address cutover that does not appear in the d230 source text. A backlog or capability map built off either AI output routes tickets to the wrong squad and schedules engineering capacity against an undocumented regulatory deadline.
What the AI got wrong, and why it matters here
Both failures inject content into customer-facing technical deliverables (release notes, schema-evolution tickets, capability map). The customer integrations team will catch the error before SaaS QA does.
Finding 1: Reconstructed stakeholder taxonomy
Opus 4.7 returned a clean stakeholder taxonomy across d224's 10 recommendations, built from category labels rather than the recommendation text. A SaaS backlog scoped off that taxonomy routes engineering work to the wrong integration squad and produces a capability map that breaks on customer architecture review.
Citation: RLB-H-INT-BIS-CPMI-API-HARMONISATION-CROSS-BORDER-2024-Q008-Opus47.
Finding 2: Fabricated November 2026 ISO 20022 cutover
Sonnet 4.6 committed to a hard November 2026 structured-address-only cutover for ISO 20022 cross-border payment messages, framed as a d230 commitment. The d230 source describes only standardisation and regulatory developments since 2023 and a separate technical annex. A schema-evolution ticket or customer release-notes entry that quotes the AI line books engineering capacity against a non-existent regulatory deadline.
Citation: RLB-H-INT-BIS-CPMI-API-HARMONISATION-CROSS-BORDER-2024-Q009-Sonnet46.
When this hits the engineering calendar
SaaS engineering and data pull CPMI material on three artefacts: the d224 capability map that drives backlog scoping, the ISO 20022 schema-evolution backlog and address-format epics, and the customer release-notes template.
| Standing item | Where the AI risk surfaces | Failure mode |
|---|---|---|
| d224 capability map | Stakeholder-to-recommendation routing | Finding 1 |
| ISO 20022 schema-evolution backlog | Cutover-date commitments | Finding 2 |
| Customer release-notes template | Both | Both |
Aggregate impact on the team
The two failures together corrupt both the capability map (Finding 1) and the schema-evolution backlog (Finding 2). The downstream cost is wasted engineering capacity and a release-notes line that contradicts the customer's own d230 read.
| Risk Impact | Count | Affected findings |
|---|---|---|
| 0 |
What this team should do
Tag the d224 stakeholder taxonomy and the d230 ISO 20022 cutover date as known-failure outputs. Any AI draft for a customer-facing technical artefact must be returned through a primary-source check before it ships externally.
Detection patterns to add to AI-review
- Stakeholder-to-recommendation mappings on d224 must be verified against the recommendation text.
- ISO 20022 cutover-date assertions against d230 must be verified against the d230 text and technical annex.
How RLB can help
RLB tracks AI failures on d224 and d230 and refreshes the catalogue against live AI subjects on rotation. SaaS engineering can wire the catalogue into the customer-deliverable review step so these two failure shapes never reach a release-notes line or a customer architecture review.