Framework lens
UK AI principles — documentation support
A documentation and evidence coverage lens for the UK's five cross-sector AI principles, with context on the sector regulators that apply them — coverage of expectations, never a legal verdict.
A documentation and evidence coverage lens for the UK's five cross-sector AI principles, with context on the sector regulators that apply them — coverage of expectations, never a legal verdict.
Why it matters
The UK is not one Act. Its approach is a set of five cross-sector principles — safety, security and robustness; appropriate transparency and explainability; fairness; accountability and governance; and contestability and redress — applied by existing sector regulators (the ICO, FCA, CMA, Ofcom, MHRA and others) rather than a single AI statute. Teams shipping AI agents that touch people in the UK want to know what those principles expect them to be able to show, and which sector regulator is likely to care, before a stakeholder asks. A documentation lens organises that and prepares structured information for your own legal and risk review — it does not stand in for it.
What good looks like
The same agent viewed through the UK principles as a coverage map, not a grade. Each of the five principles has documentation and evidence expectations mapped back to the same evidence register, controls, and tests. Higher-impact uses raise a legal-review flag. The five-principles view then helps you judge which sector regulator is likely to care — for example the ICO for personal-data handling, the FCA for a financial-services use, the MHRA for a medical-device context — as a direction for your own review, not a determination and not an automated pointer. Contestability and redress is treated as a first-class principle: whether a person affected by the agent has a documented route to challenge or appeal is visible, not assumed. Unknowns stay visible and can hold a principle at 'Not enough information' rather than being read as satisfied.
What can go wrong
A tool claims an agent is aligned with UK AI regulation as if there were one Act to align to. A sector regulator that clearly applies — the ICO, the FCA — is never surfaced, so the review that matters is missed. Contestability and redress is skipped because no single principle forced the question. Unknowns are silently treated as satisfied. The mapping is mistaken for legal advice or a regulator's view.
What AgentProof checks
AgentProof runs one deterministic, seed-driven lens engine across all profiles; the UK principles differences live in seed data pinned to a named framework version. It produces documentation and evidence coverage per principle plus an applicability read (may apply, likely relevant, not enough information, not currently relevant) and fires legal, compliance, security, and privacy review flags on higher-impact domains. It is coverage-not-compliance and reads-not-rescore: it never recomputes the readiness score and makes no legal conclusion or regulator determination. Your legal, privacy, and risk teams remain responsible. AgentProof does not provide certification, an audit verdict, or legal advice.
Key terms
The exact vocabulary this part of the record uses — grounded in the shipped product model.
- Five principles
- safety-security-robustness, transparency-explainability, fairness, accountability-governance, contestability-redress.
- Sector regulators
- ICO, FCA, CMA, Ofcom, MHRA and others — the existing regulators that apply the five principles in their sectors. AgentProof flags higher-impact uses for your own legal review; it does not identify or point to a regulator.
- Not one Act
- The UK applies cross-sector principles through existing regulators, not a single AI statute.
- Applicability
- may_apply, likely_relevant, not_enough_information, not_currently_relevant.
- Framework version
- The UK AI principles seed is pinned to a named framework version.
Keep reading
Build this record for your own agents
AgentProof turns each part of this into one documented, evidence-backed record — before your compliance review.
Ready to build a compliance-readiness record for your own agent?
Request a compliance readiness pilot to apply this guidance to a real agent.
AgentProof builds a compliance-readiness record, not an official audit, and it does not speak on behalf of any vendor.