UK open finance infrastructure needs testable governance models

At a glance

What this is: This is a research collaboration on testable infrastructure models for UK open finance, with a focus on interoperability, consent, governance, and the identity and trust mechanisms needed to make multi-sector data sharing workable.

Why it matters: It matters because open finance depends on identity, accreditation, and consent controls that can be tested across ecosystems, not just defined on paper, and those choices shape participation, resilience, and liability.

👉 Read Raidiam's research announcement on UK open finance infrastructure models

Context

Open finance is not just a data-sharing problem. It is an identity, trust, and governance problem because every participating organisation needs a reliable way to establish who can connect, what they can access, under what consent, and with what liability. The primary challenge is moving from isolated open banking patterns to shared infrastructure that still behaves predictably across sectors.

This research programme is about whether those controls can be made testable inside an FCA sandbox rather than assumed in policy documents. For IAM, IGA, and trust-and-consent teams, the key question is whether accreditation models, consent representation, and standards governance can be engineered as operational controls instead of aspirational design principles.

Key questions

Q: How should organisations govern identity in open finance ecosystems?

A: They should treat identity as an ecosystem control plane, not a single onboarding event. That means defining how participants are issued credentials, accredited, reviewed, suspended, and revoked, with those stages mapped to enforceable policy and audit trails. Without lifecycle governance, interoperability scales faster than trust and accountability.

Q: Why do consent models fail in multi-sector data sharing?

A: Consent models fail when they are descriptive but not enforceable. If one participant can interpret consent differently from another, or if the policy cannot be translated into a consistent access decision, the ecosystem accumulates governance debt. Consent must be machine-readable, auditable, and tied to the relying party that acted on it.

Q: What breaks when open finance uses inconsistent trust anchors?

A: Inconsistent trust anchors break assurance across the ecosystem. Systems may connect technically, but participants cannot reliably validate one another’s authority, which weakens onboarding, revocation, and incident response. The result is fragmented trust, where each party must build compensating controls for what should have been a shared foundation.

Q: Who is accountable when an open finance model misroutes data or consent?

A: Accountability should be pre-defined across issuer, recipient, scheme operator, and platform owner before implementation begins. If responsibility is not mapped in advance, incident handling becomes a dispute about ownership rather than a containment exercise. Governance models need liability boundaries as much as API specifications.

Technical breakdown

Identity frameworks and accreditation models in open finance

Open finance infrastructure needs a way to prove that a participant, software client, or ecosystem actor is authorised to operate inside a shared trust framework. In practice that means identity frameworks and accreditation models must bind technical access, organisational approval, and ecosystem eligibility together. If those layers drift apart, the ecosystem may scale while the trust model fragments. The article points to this as a design choice, not a paperwork exercise, because identity is what determines who can join, connect, and act across sectors.

Practical implication: define how participant identity is issued, verified, and revoked before expanding interoperability across sectors.

Consent representation and trust anchors for ecosystem data sharing

Consent representation is the machine-readable expression of what a consumer has agreed to, while trust anchors are the entities or certificates that other systems rely on to validate that agreement. In open finance, both have to survive cross-sector variation, API integration, and different legal interpretations of data use. If consent is vague or trust anchors are inconsistent, the ecosystem may technically connect but fail governance tests. This is where IAM and data governance meet: access is not enough unless consent can be enforced and audited end to end.

Practical implication: map consent artefacts to enforceable access decisions and audit them against the trust anchor chain.

Centrally coordinated, federated, and hybrid interoperability models

The research compares three infrastructure patterns: centrally coordinated models, federated interoperability, and hybrid approaches. Each distributes control differently across standards, identity, and liability. A central model can simplify consistency but may concentrate governance risk. A federated model can preserve local autonomy but increases reliance on shared rules and mutual assurance. Hybrid models often become the practical compromise, but only if the integration dependencies and operational boundaries are explicit. The value of the work is that it treats architecture as something to test, not assume.

Practical implication: evaluate each model against operational dependencies, not just policy intent, before selecting a route to scale.

NHI Mgmt Group analysis

Testable trust infrastructure is the real open finance gap: the central problem is not whether data can move, but whether identity, consent, and liability can be made testable across many institutions. Open finance fails if each participant interprets trust differently. The programme correctly shifts the discussion from abstract governance to architecture that can be simulated, compared, and challenged in a sandbox. Practitioners should treat ecosystem trust as an engineering property, not a policy slogan.

Identity and accreditation are now ecosystem-scale control points: open finance expands the attack surface and the operational blast radius of weak participant governance. When one onboarding model or trust anchor is inconsistent, the failure propagates across the network rather than remaining local. That makes accreditation lifecycle, assurance, and revocation core controls for regulators and scheme operators. Practitioners should design participant identity as a governed lifecycle, not a one-time approval event.

Consent without enforcement becomes governance debt: consumer permission only matters if it can be translated into technical decisions that every relying party must respect. That creates a direct intersection with IAM, policy enforcement, and auditability, especially where multiple sectors and processors interpret the same consent differently. Open finance programmes need a shared language for consent representation or they will accumulate invisible risk. Practitioners should make consent portability and audit traceability design requirements, not downstream fixes.

Hybrid interoperability will dominate because it reflects reality, not purity: the likely future is neither fully centralised nor fully federated, but a negotiated mix that balances control, resilience, and scale. That means architecture governance will matter as much as technical integration, because the line between platform rules and participant obligations must remain explicit. The named concept here is ecosystem trust fragmentation: when shared participation exists without shared assurance, the programme becomes harder to govern. Practitioners should map where assurance truly lives before choosing the operating model.

Agentic AI will force the next governance question: once AI starts participating in data-sharing workflows, the sector will need to decide how machine-driven actions inherit trust, consent, and liability constraints. The article correctly flags agentic AI as a future architectural dimension rather than a side feature. That is a warning for open finance leaders who assume today’s human-centric governance will cover tomorrow’s automated decision paths. Practitioners should begin defining how AI-mediated actions will be authenticated and bounded.

What this signals

Ecosystem trust fragmentation: open finance programmes fail when identity, consent, and liability are treated as separate workstreams. The practical task is to turn them into a single governable chain that can survive multi-party testing, audit, and change management. Where that chain breaks, the issue is not just interoperability, but unenforceable accountability.

The first programme signal is to expect more scrutiny of participant accreditation, trust anchors, and consent portability as open finance moves beyond design into scale. Teams that already manage non-human credentials, policy enforcement, or federation will recognise the pattern: shared access only works when the control boundaries are explicit. Identity governance becomes the testing framework for ecosystem credibility.

Open finance leaders should also plan for AI-mediated participation. The article’s mention of agentic AI is a clue that future ecosystem design will need to govern machine actions as well as human approvals, which brings identity assurance, consent inheritance, and liability mapping into the same operating model.

For practitioners

• Define participant identity lifecycle controls Specify how ecosystem participants are onboarded, accredited, reviewed, suspended, and revoked across the full open finance lifecycle. Make the identity framework explicit enough that each stage can be tested inside a sandbox against failure scenarios and policy exceptions.
• Translate consent into enforceable policy Represent consent in a machine-readable form that can drive access decisions, logging, and revocation across sectors. Ensure the consent model survives multi-party interpretation and can be audited against the relying party that acted on it.
• Stress-test interoperability against liability boundaries Map where operational responsibility shifts between scheme operator, relying party, identity issuer, and data recipient. Use that map to test whether an integration pattern creates ambiguous liability or clean accountability under incident conditions.
• Evaluate hybrid models against assurance dependencies Do not choose between centralised and federated models on principle alone. Compare the assurance dependencies, trust anchor management, and revocation handling required by each approach before committing to a production design.

Key takeaways

• Open finance becomes a governance problem when shared data needs shared trust, not just shared APIs.
• Identity, consent, and liability must be testable together or the ecosystem will scale faster than its controls.
• Practitioners should treat participant accreditation and consent enforcement as lifecycle controls, not one-off design decisions.

Key terms

• Trust anchor: A trust anchor is the root authority that signs federation metadata and establishes the policies other participants inherit. In practice, it controls who can join, what cryptographic rules apply, and how trust is delegated across an ecosystem. The security posture of the whole federation depends heavily on this layer.
• Accreditation Model: An accreditation model defines how a participant is approved to join and operate within an ecosystem. It usually combines technical assurance, policy compliance, and organisational eligibility, and it matters because weak accreditation turns interoperable access into unmanaged trust.
• Consent Representation: Consent representation is the machine-readable way a system records what a user or organisation has permitted. In shared-data ecosystems, it must be precise enough to drive access decisions, auditable enough to prove compliance, and durable enough to survive multi-party processing.
• Federated Interoperability: Federated interoperability is a model where independent organisations share data and services through common rules instead of a single central platform. It preserves local control while increasing the need for mutual assurance, consistent identity handling, and clear revocation paths.

What's in the full report

Raidiam's full article covers the operational detail this post intentionally leaves for the source:

• The research design and how the FCA Smart Data Accelerator will test infrastructure models in the Digital Sandbox
• The alternative architecture patterns under comparison, including centrally coordinated, federated, and hybrid approaches
• The governance and liability considerations that will be examined across open finance and smart data use cases
• The implementation perspective from a regulated ecosystem operator on how testable blueprints are built

👉 Raidiam's full article covers the FCA-backed research scope, architecture dimensions, and sandbox testing approach.

Deepen your knowledge

The NHI Foundation Level course, the industry's only accredited NHI security programme, covers NHI governance, machine identity security, and secrets management in a practitioner-focused format. It helps security and identity teams build the governance foundations needed for modern shared-access programmes.