TripleGRC Platform

Your management systems were designed for risks you can predict. What about the ones you can't?

178 controls across ISO 27001, ISO 27701, and ISO 42001 — modelled as complex adaptive systems, stress-tested under deep uncertainty, powered by agent-based simulation and local LLM integration. One platform, three management systems, 15 standards.

0
Controls modelled
0
Management systems
0
Standards covered
0
ABM models
01

The checklist problem

Traditional GRC treats controls as independent items to be individually assessed and audited. Risk assessment uses likelihood x impact matrices that produce inconsistent rankings and miss compound failures. The same methodology is applied identically to information security, privacy, and AI governance — despite fundamentally different uncertainty structures.

Scenario discovery doesn't estimate the likelihood of threats. It identifies the conditions under which your defences fail — regardless of which threat materialises.
02

Three management systems, one analytical framework

We apply Decision Making Under Deep Uncertainty (DMDU) — the same methods used by RAND, Deltares, and the World Bank for climate adaptation and infrastructure planning — to security, privacy, and AI governance.

ISMS | ISO 27001

Information security

93 Annex A controls modelled as a coupled system. Four coupling types reveal how training failure cascades into malware susceptibility, how budget pressure creates correlated control degradation.

93 controls | 400+ couplings | 4 types
4 sector clusters | 3 ABM models
+ 27005 | 27007 | 27035 | 27799
Explore ISMS coupling →
PIMS | ISO 27701

Privacy management

47 privacy controls with 9 privacy-specific coupling types. Consent chains, rights cascades, cross-border transfer risk. The controller/processor split creates accountability gaps that traditional auditing misses.

47 controls | 160 couplings | 9 types
4 privacy clusters | 3 ABM models
+ 27557 | 29100 | 29134 | 31700
Explore PIMS coupling →
AIMS | ISO 42001

AI governance

38 AI controls with coupling types unique to AI: data dependency, lifecycle sequencing, risk amplification, governance tension. The system under management is itself adaptive — governance must be too.

38 controls | 163 couplings | 6 types
4 AI clusters | 3 ABM models
+ 38507 | 23894 | EU AI Act | NIST AI RMF
Explore AIMS coupling →

62 cross-references bridge the three systems. Investing in bridge controls simultaneously strengthens all three management systems.

03

Map → explore → discover

Step 1

Coupling analysis

Map directed interactions between controls. Information flows, failure propagation, consent chains, data dependencies. The coupling structure reveals which controls fail together.

Step 2

Community detection

Label propagation identifies tightly coupled control subsystems. Each community is scored for ABM candidacy based on coupling density and cross-category diversity.

Step 3

Agent-based simulation

10-phase NetLogo engine, 18 parameters, no hardcoded rates. Five feedback loops produce cascading failure, workaround contagion, normalisation of deviance, and recovery.

Step 4

Scenario discovery

Pre-configured BehaviorSpace experiments generate parameter sweep data. The DMDU pipeline feeds into PRIM and CART analysis to isolate the specific combinations — budget-pressure x turnover x threat-novelty — that drive system failure.

04

Platform capabilities

13 pages

Interactive analytical tools

Coupling discovery for ISMS, PIMS, AIMS. Connectivity analysis. Cluster identification. NetLogo export. ABM simulation. Standards mapping for 15 standards.

9 models

Pre-built NetLogo simulations

3 ISMS (training, perimeter, data-regulated), 3 PIMS (consent, rights, high-sensitivity), 3 AIMS (oversight, data quality, transparency). BehaviorSpace experiments pre-configured.

Client engine v3

Engagement management

Scope selector (ISMS/PIMS/AIMS). Structured consultant notes: diagnosis, document review, session log, running findings with severity classification. JSON/CSV import/export.

LLM integration

Local AI assistant

Connects to Ollama (qwen3:4b recommended). Document ingestion: upload client docs → extract profile. Report generation: analysis results → tailored narrative. All local — no data leaves your machine.

05

Standards coverage

SystemCoreControlsCoupling typesExtended standards
ISMS2700193427005, 27007, 27035, 27799
PIMS2770147927557, 29100, 29134, 31700
AIMS4200138638507, 23894, EU AI Act, NIST AI RMF
06

Reports & deliverables

ReportSystemAudience
Service OverviewISMSProspects
Technical DocumentationISMSEngaged clients
User GuideISMSEngaged clients
Privacy Service OverviewPIMSProspects
Privacy Technical GuidePIMSEngaged clients
AI Governance Service OverviewAIMSProspects
AI Governance Technical GuideAIMSEngaged clients
SSRN Working PaperAllAcademic / thought leadership

Plus LLM-generated custom reports: findings & recommendations, executive summaries, technical analysis, implementation roadmaps, risk assessment narratives — all tailored to the specific client profile and analysis results.

Start exploring

Pick a management system to explore its coupling structure, or load a client profile to begin an engagement.

ISMS coupling → PIMS coupling → AIMS coupling → Client profile → LLM assistant →