The AI Race for Evidence-Grade Data at Scale

Guest Blog: Co-Founders of Df2020, John Rawlings and Frederick McMahon

Systemic Weaknesses Hidden in Procedural Execution

Society runs on complex procedures yet possesses remarkably weak evidence-grade visibility into how those procedures are actually executed in practice. As complexity, velocity, and volatility accelerate, this widening gap between governance and operational reality is fracturing the resilience of conventional Governance, Compliance, and Risk operating models.

Why Procedural Complexity Escaped Automation

Where procedural execution could be automated, organisations largely automated it. What remained was a growing landscape of complex human procedures that lacked the structured execution data needed for meaningful automation.

The Absence of Evidence-Grade Execution Data

The inability to capture evidence-grade data for each step of procedural execution creates fertile ground for hidden risks, rework, delays, rising operational costs, and growing governance liabilities. As a result, operational reality is often reconstructed retrospectively through reporting, interpretation, audits, surveys, spreadsheets, and administrative controls.

The underlying execution signals can therefore remain incomplete, delayed, misleading, or weakly connected to how procedural execution actually occurred.

Compensating for Weak Procedural Visibility

Organisations have historically compensated for this weakness through increasing layers of supervision, assurance, compliance, auditing, governance oversight, management control, and support from Subject Matter Experts.

These overheads materially increase operational costs. Yet hidden operational and regulatory risks can still emerge due to the continued absence of evidence-grade execution data.

Time to Think Differently

Procedures are the means through which policies are implemented across operational systems in the public and private sectors. Their importance is grounded in the way we govern and operate. Yet despite governing the operational execution of modern society, complex procedures have never been treated as a strategic transformation priority across successive waves of technology.

Unless we fundamentally rethink how complex procedures are operationalised and governed, accelerating complexity, velocity, and volatility will drive growing systemic fragility across local, national, and international systems.

Human Control in an Age of Accelerating Complexity

Unlike Generative AI, Symbolic AI allows humans to remain in direct control of how procedural knowledge is structured, narrated, governed, measured, audited, and evolved. As the pace of change accelerates, maintaining human control over procedural knowledge becomes increasingly important as the gap between governance and operational reality continues widening.

Symbolic AI Needs a Next Generation

Traditional Symbolic AI focused heavily on structured rules, structured data, and procedural automation. However, many of the most complex procedural environments could not be meaningfully automated because the underlying execution data did not exist in sufficiently structured form.

The next generation of Symbolic Agentic AI therefore shifts from automating procedures toward governing procedural execution through structured human interaction capable of continuously generating evidence-grade operational intelligence.

This requires procedural knowledge to evolve through modular, continuously governable ecosystems rather than monolithic procedural systems.

What Is a Symbolic Agentic AI Ecosystem

At the centre of a Symbolic Agentic AI Ecosystem sits a multi-dimensional ontology governing how procedural knowledge, decisions, actions, risks, controls, outcomes, and auditability are mapped and related. Loosely coupled agents evolve independently while remaining continuously governable through the shared ontology.

Procedural complexity and permutations of execution can scale without collapsing into monolithic procedural systems, while masking underlying complexity from the individual carrying out the work.

Real-Time Evidence-Grade Governance Intelligence

Through governed human interaction, procedural execution becomes measurable, auditable, continuously evolvable, and capable of generating evidence-grade governance intelligence in real time.

Procedural execution continuously generates Key Risk Indicators (KRIs), Key Performance Indicators (KPIs), real-time audits, and earlier visibility into hidden operational and regulatory risks. Governance can therefore increasingly shift from retrospective interpretation toward direct evidence-grade visibility into procedural execution itself.

Strategic Benefits

This materially reduces the layers of supervision, assurance, compliance, auditing, governance oversight, management control, and Subject Matter Expert dependency that can increase operational costs by an estimated 15–25%.

As procedural execution becomes continuously measurable and auditable, organisations can become significantly more agile, adaptive, and resilient in responding to accelerating operational, regulatory, and market change. Reduced dependency upon tribal knowledge increases workforce flexibility and accelerates onboarding across complex operational environments.

Operational and regulatory risks can be identified in real time before escalating into reputational damage, litigation, operational disruption, or balance-sheet exposure. Forensic investigations can be conducted far more rapidly. Procedures can also be enriched, adapted, and extended far more quickly without the coordination costs traditionally associated with complex procedural change.

A New Strategic Frontier for AI

The global AI race is increasingly shifting toward access to high-quality proprietary data capable of generating strategic advantage beyond publicly available internet data. Policies, standards, regulations, and best practices already contain vast amounts of procedural knowledge embedded across the public and private sectors.

The problem is not weak knowledge. The problem is weak evidence-grade data representing how this knowledge is actually executed in practice.

Symbolic Agentic AI Ecosystems continuously generate proprietary evidence-grade data in real time at scale. Every shared procedural knowledge asset becomes a scalable production system for continuously generating proprietary evidence-grade data across multiple organisations, sectors, and operational environments.

As this evidence-grade intelligence compounds, benchmarking, operational learning, governance resilience, and procedural optimisation can continuously evolve across entire sectors and ecosystems. The nations and organisations that first operationalise procedural execution as a continuously measurable and governable intelligence layer will secure profound strategic advantages in the future development of AI-driven economies and societies.