Articles.
Essays that need more room than a post. The list is not a feed.
Agentic AI and SM&CR Accountability
When an agent acts for a bank, the Senior Managers Regime still names a person. The Mills Review records that firms have asked for a clearer account of reasonable steps. Riley Betts would want the FCA’s forthcoming examples of good and poor practice to show what that account contains.
Read →Riley Betts awarded SMART: SCOTLAND funding for the Compliance Operator Framework
Riley Betts has been awarded SMART: SCOTLAND funding from Scottish Enterprise for the Compliance Operator Framework, a feasibility project for governed, auditable AI workflows in regulated financial environments.
Read →AI governance under the EU AI Act: why evidence shape matters more than documentation volume
The EU AI Act does not reward thicker model cards. High-risk systems need records shaped for examination: which rule held, on which action, at which time — not a narrative assembled after the fact.
Read →What zero-knowledge proofs mean for regulated AI workflows
A zero-knowledge proof lets you show that a compliance rule held without handing over the underlying client data. Examination and confidentiality need not be a trade-off.
Read →Agentic AI in financial services: the architectural separation that makes compliance possible
In wealth, payments, and capital markets, agents will propose actions inside mandates and rails you already own. Compliance holds if a validator decides — not if the model is usually right.
Read →What the Riley Betts Proof Fabric does that the rest of fintech still can’t
Copilots, messaging upgrades, human-in-the-loop, GRC monitoring, stablecoins, and public-chain books each miss a different part of the same question. The Riley Betts Proof Fabric sits in the gap they leave.
Read →Why AI governance evidence must exist at decision time, not reconstructed after
Log archaeology is not evidence. When a regulator asks why an action was allowed, the record needs to have existed at the moment of decision—not assembled from scattered systems the week before the examination.
Read →The 0.1% problem: why probabilistic AI cannot govern itself in regulated workflows
An AI that follows compliance rules 99.9% of the time violates them 0.1% of the time. In a high-volume regulated workflow, that is not a rounding error. Probability is not a defence.
Read →How to prove an AI agent followed the rules: a technical architecture guide
The difference between a log that says what happened and a proof that shows the rule held before the action ran. Architecture, not monitoring.
Read →Formal specification vs. iterative debugging: what changes when you specify before AI generates
Testing tells you what happened in the cases you thought to test. Formal specification tells you what can happen across every possible input. In regulated AI, the difference is not academic.
Read →