Appearance
Portfolio Overview
In Modulos, risk quantification is a primary part of AI governance. It changes the question "what could go wrong" into a monetary value. You can use this value to set priorities and budgets, to delegate decisions, and to do audits.
Governance loop
Four stations, one operating model.
Continuous risk governance
Quantification continues for the full life of an AI System.
Identify
Give each risk a clear name that all teams use
Quantify
Change threats into a monetary expected loss
Mitigate
Do mitigations that have a measured effect
Monitor
Quantify again when systems and data change
The dashed arc marks restart — every cycle re-enters Identify with what changed since the last pass.
What this is
Risk quantification does not remove risk. Quantification makes risk clear. Leaders can then select which risks to accept, which risks to decrease, and which activities to stop.
When risk has only a qualitative rating, governance can become a veto without a basis. When risk is a monetary value, the decision uses facts: the expected loss, the risk appetite, and the cost of mitigation.
Quantification also makes decisions quicker:
- decisions at each gate of the AI System life cycle become clear and repeatable
- teams can make decisions in the limits that the organization sets
- "too risky" becomes a statement that you can examine and make better
Where in Modulos
Use these tabs and pages to see and manage the risk exposure:
Organization → Risk Management → Risk Overviewfor the totals, the trend, and the risks with the highest valuesOrganization → Risk Management → Portfoliofor a map of the risk values against the limits, and for the economic value of each AI System against its risk valueOrganization → Risk Management → Risk LimitsandAI System Risk Limitsfor the appetite and the limitsAI System → Risksto add risks, select threats, and quantify
For the roles and the rules for totals, see Risk Operating Model.
Permissions
- Risk Managers change the risk configuration of the organization.
- AI System Owners usually start the quantification runs.
- On the Portfolio tab, users without full access to all AI Systems see only the AI Systems where they have a role.

- 1Total Organization Risk ExposureThe sum of the risk values of all AI Systems, with the limit of the organization.
- 2Risk Distribution by CategoryThe value of the first five categories, each against its limit.
- 3Top 3 Highest Risk Value AI SystemsThe AI Systems with the highest values. Select one to open its Risks page.
The Risk Overview tab
The Risk Overview tab in Organization → Risk Management is the dashboard of the organization. It shows these items:
| Element | What it shows |
|---|---|
| Total Organization Risk Exposure (KPI) | The sum of the risk values of all AI Systems. The Limit below the value is the total risk appetite. |
| Total Economic Value (KPI) | The sum of the annual economic values of the AI Systems. |
| Risks Identified (KPI) | The number of risks in the organization taxonomy. This number includes the archived risks. |
| Risk / Total Appetite (KPI) | The total risk appetite of the organization, as money. |
| Risk Distribution by Category | The value of the first five categories, each against the limit of the category. |
| Top 3 Highest Risk Value AI Systems | The three AI Systems with the highest values, each against its limit. Select one to open its Risks page. |
| Risk Value Over Time | The total risk value across time, and a line for the total risk appetite. The Stacked and Breakdown views show each AI System. |
| Top 3 Highest Risk per Category (across AI Systems) | The three risks with the highest values in each category, each with the name of its AI System. |
The Risk Overview tab counts only AI Systems with the registration status Pending or Approved, and not the organization AI System. The values follow the rules in How risk becomes a portfolio signal. To change the appetite or a category limit, use the Risk Limits tab.
Organization history
The organization has no Quantification History table. The history of the runs is on the Threat Quantification Detail page of each threat. See Run history and trends.
How it works
In Modulos, the value of a risk comes from its threats:
text
Expected loss = Σ (rate × damage)- Rate: how frequently the threat becomes an incident.
- Damage: the monetary loss for each incident.
This division into rate and damage shows where the risk comes from. It also shows which part a mitigation must decrease.
Platform objects and rollups
- Organization taxonomy: shared categories, risks, and threat vectors.
- AI System risks: taxonomy risks that an AI System uses.
- Risk threats: threat vectors in one AI System.
- Quantification runs: the method, the inputs, the outputs, the status, and one monetary value for each risk threat.
- Rollups and limits: risk threat → risk → AI System → category → organization.
Risk model
Three bands of structure, one rollup column.
Organization
Threat vectors
Risks
Categories
Project
Project risks
Risk threats
Quantification
Quantification runs
€Expected loss
Rollup
The aggregation column
Quantification happens at the threat level. Only quantified threats contribute to totals; everything else is a sum.
threat
→ risk
→ project
→ category
→ organization
Limits apply at every layer the column touches: risk, project, category, organization. Threats are where the numbers come from; everything above is aggregation.
The Portfolio tab
The Portfolio tab has two views. Select Treemap or Value and Risk at the top of the tab.
- Treemap shows the risk values as a map of tiles and compares them with the risk limits. See Treemap.
- Value and Risk compares the annual economic value of each AI System with its risk value, and gives each AI System an assessment. See Value and Risk.
Each user of the organization can open the Portfolio tab when risk management is on for the organization. The two views include only AI Systems with the registration status Pending or Approved, and not the organization AI System. Users without full access to all AI Systems see only the AI Systems where they have a role.
Risk appetite and limits
Limits connect quantification to the risk appetite of the organization:
- The organization sets a total monetary risk appetite.
- Each category gets a percentage of the appetite.
- Each AI System with the status Pending or Approved, other than the organization AI System, gets a risk limit from the appetite.
- In each AI System, each risk gets a limit from the budget of its category.
Risk budget cascade
One appetite, four layers, one conservation rule.
Organization
€15M
Categories
Technical€3M
Operational€6M
Legal & Compliance€3M
Ethical€1.5M
Governance€1.5M
Projects
LLM chatbot€2.4M
HR detection€2.0M
Pricing model€1.6M
Risks
Human–AI failure€900K
Insufficient testing€700K
Lack of transparency€800K
Conservation
Budgets must add up
The appetite is a single number. Every layer below must sum back to it; otherwise quantification is blocked.
Σ Category shares = 100%
Σ Project limits = appetite
Σ Risk limits = category
When sums fail, Modulos surfaces a warning and blocks quantification on the affected scope. The conservation rule is the same at every layer — only the entities change.
For the rules and the warnings, see Risk limits.
Portfolio reality
The sum of the expected losses is a good start. But shared vendors, shared foundation models, and shared infrastructure can connect risks. For exposures that collect in one area, use scenario analysis and examine the dependencies.
How to use it
1
Set appetite
Set a total risk appetite and the category limits
2
Adopt the taxonomy
Start with the default library and add the risks of your domain
3
Quantify the top threats
Quantify 1–3 threats for each important risk
4
Mitigate and re-quantify
Do mitigations, then quantify again when systems and data change
Related pages
Treemap
A map of the risk values against the risk limits
Value and Risk
Compare the economic value of each AI System with its risk value
Organization Taxonomy
A shared library of categories, risks, and threat vectors
AI System Risks
Add taxonomy risks to an AI System
Risk Quantification
Get monetary values and make your estimates better from run to run
Acting on Quantified Risk
Compare values with limits, decrease risk through Controls, and quantify again