AIHood
MII-29 · Machine Insurability Index

The 29 identities that make a robot insurable

MII-29 is the spine of verifiable digital identities a robot must expose to enter insurance, procurement, and liability governance. Grouped into five clusters, each identity answers a question an underwriter, buyer, or regulator has to settle before a robot can be covered, admitted, or held accountable.

ISO, GB, and UL make robots safer; MII-29 makes them verified, priced, and accountable.

Cluster 1 of 5

Machine

The physical unit and its hardware provenance — proving which machine, which core parts, and which sensors and actuators were present.

Machine-body identity

01
Problem it solves
Is this the same physical machine it claims to be?
Insurance meaning
Confirms the exact object under the policy.
Identity

Hardware root-of-trust

02
Problem it solves
Is the identity bound to the hardware, not just software?
Insurance meaning
Prevents swapping and forgery of the insured unit.
Identity
Cyber Resilience

AEM / evidence-module identity

03
Problem it solves
Which module generates the robot's evidence, and can it be trusted?
Insurance meaning
Establishes the provenance of incident evidence.
Audit Logs

Main-controller SoC identity

04
Problem it solves
Has the core control unit been swapped or altered?
Insurance meaning
Flags changes to the part that governs behavior.
Cyber Resilience

Sensor identity

05
Problem it solves
Can the robot's perception data (camera, LiDAR, IMU, force) be trusted?
Insurance meaning
Grounds the trustworthiness of sensor evidence.
Audit Logs

Actuator identity

06
Problem it solves
Is each physical action traceable to a specific joint, motor, or gripper?
Insurance meaning
Assigns liability for the action a part took.
Operational Safety

Battery / BMS identity

07
Problem it solves
Can the battery and its management system be trusted thermally?
Insurance meaning
Prices fire, thermal-runaway, and downtime risk.
Operational Safety

Comms-module identity

08
Problem it solves
Is the robot's communication link (remote, wireless, 5G, Wi-Fi) trusted?
Insurance meaning
Enables attribution of cyber-attacks over the link.
Cyber Resilience
Cluster 2 of 5

Makers

Who built, integrated, and supplied the unit — the chain of product, whole-unit, supply-chain, and software liability.

Manufacturer identity

09
Problem it solves
Who actually built the robot?
Insurance meaning
Anchors product liability.

OEM / integrator identity

10
Problem it solves
Who integrated the parts into a whole unit?
Insurance meaning
Anchors whole-unit liability.

Component-supplier identity

11
Problem it solves
Which key parts came from which supplier?
Insurance meaning
Enables supply-chain recourse after a defect.

Software-vendor identity

12
Problem it solves
Whose software controls the robot?
Insurance meaning
Anchors software liability.
Cyber Resilience
Cluster 3 of 5

AI

Model, data, and runtime-config provenance — whose model ran, which version, on what data, and under which parameters.

AI-model-provider identity

13
Problem it solves
Whose AI model is driving the robot?
Insurance meaning
Anchors model liability.

Model-version identity

14
Problem it solves
Exactly which model version was running at incident time?
Insurance meaning
Enables replay and OTA-update liability.
Cyber Resilience

Dataset / training-source identity

15
Problem it solves
Where did the training data originate?
Insurance meaning
Surfaces bias, defect, and compliance exposure.

Deploy-config identity

16
Problem it solves
Which parameters, policies, and thresholds were active?
Insurance meaning
Attributes behavior to its runtime configuration.
Fleet Visibility
Cluster 4 of 5

People

Owner, operators, and maintainers — who owned, ran, intervened, remote-controlled, and serviced the robot.

Owner identity

17
Problem it solves
Who owns the asset?
Insurance meaning
Anchors property and asset liability.
Identity

Operator identity

18
Problem it solves
Which organization operates the robot?
Insurance meaning
Anchors operational liability.

Human-operator identity

19
Problem it solves
Who intervened manually at the moment it mattered?
Insurance meaning
Attributes a specific human action.

Remote-controller identity

20
Problem it solves
Who sent remote commands to the robot?
Insurance meaning
Assigns remote-control and cyber liability.
Cyber Resilience

Maintenance-provider identity

21
Problem it solves
Who services and maintains the robot?
Insurance meaning
Anchors maintenance liability.
Operational Safety
Cluster 5 of 5

Governance

Task and scene scope plus the underwriting and liability lifecycle — from what the robot did and where, to policy, evidence, incidents, permits, and accountability status.

Task identity

22
Problem it solves
What task was the robot actually executing?
Insurance meaning
Ties coverage and liability to the task performed.

Scene / ODD identity

23
Problem it solves
Where was it operating, and within what boundary?
Insurance meaning
Establishes operating-domain (ODD) exclusion liability.
Fleet Visibility
Operational Safety

Insurance-use-domain (IUD) identity

24
Problem it solves
What usage scope is actually insured?
Insurance meaning
Defines the policy's coverage boundary.
Compliance

Policy identity

25
Problem it solves
Which policy covers this robot right now?
Insurance meaning
Connects the unit to underwriting and claims.
Compliance

Evidence-pack identity

26
Problem it solves
Which evidence pack documents a given incident?
Insurance meaning
Supplies the claim's evidentiary record.
Audit Logs

Incident / event identity

27
Problem it solves
Which specific incident is being handled?
Insurance meaning
Enables case management of a claim.
Audit Logs

Regulatory / permit identity

28
Problem it solves
Is the robot approved to operate in this scene?
Insurance meaning
Gates government and procurement admission.
Compliance

Accountability-status identity

29
Problem it solves
What is its current liability, recall, revocation, or restriction status?
Insurance meaning
Drives renewal, recourse, and ban decisions.
Fleet Visibility