THE INTELLIGENT REGULATOR LEADERSHIP PROGRAMME
Put judgement where it belongs.
15 Sessions · 25 Minutes Per Day · Three Weeks · Free · For regulators and regulated companies in Nigeria’s petroleum industry
Everything this programme rests on, on one page, taken from the book. Print it and keep it in the folder you take to the executive review. Make the evidence reliable, put judgement where it belongs and close the loop after action.
| Framework | What it says |
|---|---|
| The Operating-Intelligence Loop | Observation, interpretation, decision, action and learning. In a licensing process: receive the application and identify its evidence; determine the relevant provisions and criteria; the authorised officer or institution decides; communicate the determination and manage its conditions; compare the result with later queries, appeals and inspection findings. The loop fails wherever responsibility disappears. Before sponsoring a project, finish the sentence: this evidence will help this person make this decision by this time, leading to this action and this measurable outcome. |
| The Data-Foundation Precondition | Dependable judgement needs dependable evidence: the identity of the facility, the period covered, the units, the source, the revision, the person responsible and the limitations. AI does not remove this precondition; it often makes weak data more dangerous. The first data contract should be small enough to enforce. |
| The Autonomy Ladder | Assistance that retrieves or organises evidence; supervised drafting only after source fidelity is demonstrated; bounded execution for low-consequence tasks when the permission boundary, validation and recovery procedure work. Legally consequential determinations stay with authorised human decision-makers unless a lawful and specifically approved arrangement establishes otherwise. The ladder is not a timetable: autonomy is earned by the action and its consequences. |
| The Value-versus-Risk Map | Compare the expected public or operational benefit with the harm of an error: the people affected, the opportunity to correct, the time available, confidentiality and the capacity to recover. Describe the controls in concrete terms rather than assigning a reassuring colour. |
| The Sequencing Discipline | Sequence around dependencies: current legal sources before legal retrieval; facility identities before cross-system analytics; inspection findings and closure records before a risk-ranking model; decision rights before an agent with tools that affect external parties. Some early improvements will use conventional software, and belong in the programme. |
| The Operations-Excellence Lens | Examine the whole working process: where the case waits, who must supply missing evidence, what happens when the portal is unavailable. Trace one application, one inspection and one complaint from receipt to disposition. A model that improves one task while increasing downstream rework has not improved the operation. |
These six frameworks are author tools, not statutory standards. The book is not an official interpretation issued by a regulator.
Use the frameworks together on a single decision. Record the purpose, legal and technical basis, evidence, uncertainty, accountable owner, permitted AI contribution, required human checks, action and review date. Preserve the original record and each material correction. The institution should be able to explain the decision to a reviewer who was not present at the meeting.
| Activity | AI contribution | Accountable human responsibility |
|---|---|---|
| Index a submission | Propose identifiers and evidence links | Case officer verifies identity and scope |
| Retrieve a provision | Return source and relevant passage | Qualified reviewer establishes applicability |
| Assess technical evidence | Identify candidate inconsistencies | Competent professional assesses adequacy |
| Prioritise supervision | Propose rank with evidence and limits | Authorised supervisor sets priorities |
| Draft a query | Prepare source-linked working text | Responsible officer verifies and authorises |
| Issue a determination | Assemble verified supporting record | Lawful decision-maker determines and issues |
| Close a finding | Compare response with requirement | Authorised reviewer verifies closure |
| Change the AI service | Support comparison and evaluation | Service owner controls change and approval |
An author matrix: adapt it to actual lawful authority and organisational policy. Job titles alone do not establish delegated powers.
| Period | Main work |
|---|---|
| Days 1 to 15 | Establish the decision: trace representative cases, set the baseline, confirm legal sources and decision rights, name the executive sponsor, process owner and service owner, and write the pilot charter before purchasing. |
| Days 16 to 30 | Prepare the evidence: resolve identities, revisions and access; build a controlled evaluation set with a separate final set; arrange independent reference judgements; document the workflow and fallback. |
| Days 31 to 60 | Run in shadow mode: outputs without external action, compared with the approved reference process; test poor scans, conflicting revisions, unsupported questions and hostile document instructions; measure reviewer burden as well as time saved. |
| Days 61 to 90 | Decide on controlled use: review against the acceptance gates, obtain professional and operational sign-off, verify source fidelity, access, audit, recovery, competence and sustainable cost, and state the permitted scope — or record why the pilot did not pass. |
| Months 4 to 6 | Stabilise the service under real workload and test whether the benefit is real. |
| Months 7 to 12 | Expand by dependency, only after the first service is stable. |
An author planning framework, not a legal deadline or a promise that any institution will reach maturity within a fixed period. A programme should be able to stop a weak pilot without calling that responsible decision a failure of ambition.
| Measure | Meaning | Interpretation limit |
|---|---|---|
| Median case age | Middle elapsed age in a defined population | Does not reveal oldest difficult cases |
| Tail case age | Age at a stated percentile or oldest material case | Needs consistent case scope |
| Avoidable rework rate | Cases with verified process-caused repetition divided by reviewed cases | Requires agreed cause classification |
| Source fidelity | Verified material assertions supported by the correct source | Sample and consequence matter |
| Open material actions | Findings or conditions without verified closure | Risk should remain visible |
| Correction rate | Cases requiring material workflow-related correction | Distinguish AI and other causes |
| Net capacity released | Time saved less review and service effort | Is not automatically cash saving |
| Recovery readiness | Result of a defined restore and fallback test | A policy alone is insufficient |
Author measures, not official institutional targets. Agree definitions and denominators before reporting, and claim cash savings only when an actual cash-cost change is established. The scorecard is useful when it changes action.
| # | Session | Plan section |
|---|---|---|
| 1 | The Intelligent Regulator and the Reader's Toolkit | Section 1 — Our organisation, our side of the regulatory relationship, the one decision we will carry and where its loop breaks. |
| 2 | From Regulatory Data to Public Value | Section 2 — The public-value chain, the observed baseline, three statuses kept apart and the three owners. |
| 3 | Making the PIA an Operating Backbone | Section 3 — The obligation-to-evidence chain and one Appendix B register entry, with law kept apart from implementation. |
| 4 | Building a Regulatory Evidence System | Section 4 — The evidence system: identities, provenance, the data contract and a decision we can reconstruct. |
| 5 | The Architecture of Trust | Section 5 — The architecture of trust: layers, the plant boundary, data protection, exchange events and the fallback. |
| 6 | Licensing and Approvals That Can Be Explained | Section 6 — The decision route, completeness kept apart from adequacy, an answerable query and conditions that can be monitored. |
| 7 | Risk Based Supervision and Inspection | Section 7 — Supervision and closure: the prioritisation method, the evidence boundary and verified closure. |
| 8 | Measurement and Revenue Assurance | Section 8 — The reconciliation: boundary, comparable basis, competing explanations and a benefit counted once. |
| 9 | Safety and Environmental Intelligence | Section 9 — Safety and environmental evidence: leading signals, escalation, provisional status, measurement basis and long-term obligations. |
| 10 | Host Communities and Accountable Development | Section 10 — Accountable delivery: funding, spending, delivery and service kept distinct, and consultation made inspectable. |
| 11 | Gas Markets and Downstream Consumer Protection | Section 11 — Recourse: constraint, complaint, anomaly and breach kept apart, and a response people can understand. |
| 12 | AI Governance and Administrative Fairness | Section 12 — Governance and fairness: the inventory, meaningful review, fairness tests, the acceptance case and the route to correction. |
| 13 | Enforcement, Cybersecurity and Controlled Autonomy | Section 13 — Investigation, security and controlled autonomy: the action boundary and the five acceptance gates. |
| 14 | Better Regulatory Partners and Nigerian Capability | Section 14 — The partnership and the capability: the submission and query discipline, waiting by cause, and the capability we keep. |
| 15 | The First Ninety Days: Your Signed 90-Day Regulatory AI Plan | The signed 90-Day Regulatory AI Plan. |
ERW CommandRegulator · built on The Intelligent Regulator — An Operating Doctrine for AI Leadership in Nigerian Petroleum Regulation and Industry Collaboration by Ritchie Wingo. Leadership training, not legal, investment or engineering advice, and not an official interpretation issued by a regulator. Back to the programme home →